US11306308B2 - High-throughput CRISPR-based library screening - Google Patents

Abstract

Provided herein is an improved method for performing CRISPR/Cas based screening that is not dependent on viral cloning methods.

Description

RELATED APPLICATIONS

This application is a national stage filing under 35 U.S.C. § 371 of International Application No. PCT/US2016/061691, filed Nov. 11, 2016, which was published under PCT Article 21(2) in English and claims the benefit under 35 U.S.C. § 119(e) of U.S. Provisional Application Ser. No. 62/255,359 entitled “HIGH-THROUGHPUT CRISPR-BASED SCREENING” filed on Nov. 13, 2015, the entire contents of each of which are incorporated by reference herein.

FEDERALLY SPONSORED RESEARCH

This invention was made with Government support under Grant Nos. R01 HG008754, U01 HG007037 and K01 DK101684 awarded by the National Institutes of Health. The Government has certain rights in the invention.

REFERENCE TO A SEQUENCE LISTING SUBMITTED AS A TEXT FILE VIA EFS-WEB

The instant application contains a Sequence Listing which has been submitted in ASCII format via EFS-Web and is hereby incorporated by reference in its entirety. Said ASCII copy, created on Oct. 7, 2020, is named M065670371US01-SEQ-MAT and is 55.7 kilobytes in size.

BACKGROUND

Gene regulation provides the basis for cell type-specific function, and differences in cis-regulatory DNA underlie human variation, disease, and cancer. Our rudimentary understanding of gene regulation impairs our ability to predict the effects of cis-regulatory variants on gene expression and to predictively alter gene expression during stem cell differentiation and reprogramming.

Strides have been made over the last decade to catalogue gene regulatory elements. A histone modification code has been found to correlate with cis-regulatory elements such as enhancers and promoters and states such as active and poised^1-5. Gene expression reporter assays, which can now be done in high-throughput^6-8, have confirmed elements that are sufficient to activate gene expression in heterologous contexts. Additionally, techniques to identify distal DNA interactions have begun to associate enhancers with their cognate promoters^9-12, which are often not in close proximity and can at times be megabases apart.

However, existing gene regulatory techniques have several shortcomings. Reporter assays, by focusing on elements that are sufficient to activate gene expression in a heterologous context, cannot characterize elements that are necessary but not sufficient for gene expression or elements whose activity does not transfer to a non-native context. Additionally, genes can have dozens of regulatory elements, and there is no high-throughput approach capable of determining the relative importance of each gene regulatory element on native gene expression levels.

SUMMARY

This disclosure provides, inter alia, methods for studying the function of genomic regions, as well as methods for identifying genomic regions and elements involved in transcription regulation. These methods rely in part on a newly developed CRISPR/Cas system-based library screen provided herein. This disclosure therefore also provides an improved method for performing CRISPR/Cas system-based library screens.

The method exploits the ability of the CRISPR/Cas system to cleave specific genomic sites in order to introduce exogenous guide sequences of interest into the cell and then to allow such exogenous sequences to target the CRISPR/Cas system to homologous genomic sites in the cell. The targeted genomic sequences are then mutated in the host cell, and the effect of such mutation is analyzed. As is known in the art, CRISPR/Cas system nucleases require a guide RNA to cleave genomic DNA. These guide RNAs are composed of (1) a 19-21 nucleotide spacer (guide) of variable sequence that targets the CRISPR/Cas system nuclease to a genomic location in a sequence-specific manner, and (2) an invariant hairpin sequence that is constant between guide RNAs and allows the guide RNA to bind to the CRISPR/Cas system nuclease.

The methods provided herein involve a first step of integrating one or more DNA sequences that each code for a dummy guide RNA, into the genome of a cell, at one or more typically defined loci. This step comprises integrating a defined (or controlled) number of such DNA sequences into the cell. The integration occurs at a defined (or controlled) number of loci in the cell. In some instances, the number of DNA sequences and loci will be known to the end user. There may be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more such DNA sequences integrated into the genome of the cell, and such DNA sequences may be integrated in proximity to each other or distally to each other. When a host cell has more than one dummy guide RNA construct integrated, such dummy guide RNA constructs may be identical to each other (i.e., their sequences may be identical) or they may differ from each other (i.e., their sequences may be different).

The “variable” or guide sequence of the dummy guide RNA is not homologous to any sequence in the host cell and thus it is not able to direct a CRISPR/Cas nuclease to any particular site in the genome. Thus the presence and/or expression of the dummy guide RNA, whether in the presence or absence of a CRISPR/Cas nuclease, does not trigger a CRISPR/Cas-based genomic cleavage event in the cell. Accordingly, such dummy guide RNA is referred to as defective.

The second step introduces into the host cell a library of DNA fragments that each comprises a candidate variable/guide sequence. The variable/guide sequences in these DNA fragments may be derived from genomic DNA of the cell type being used as the host cell. These fragments may be prepared through fragmentation of genomic DNA, optionally followed by amplification using any number of amplification techniques, including but not limited to PCR. Through the amplification process, the genomic fragments may be modified to facilitate their integration into the dummy RNA genomic locus. For example, the genomic fragments may have additional sequences added to one and typically both ends during amplification. These end sequences may be homologous to sequences that flank the variable/guide sequence at the integrated dummy guide RNA construct. Using a CRISPR/Cas system nuclease such as Cas9, the variable/guide sequence of the dummy guide RNA construct is cleaved and, through a process of homologous recombination, is replaced with the variable/guide sequence in one of the amplified fragments to form a new guide RNA construct or locus. The new guide RNA that is encoded by the recombined locus is homologous to a genomic region in the cell, and it is thus able to direct a CRISPR/Cas nuclease to that specific genomic region in the cell, thereby facilitating cleavage of that region by the CRISPR/Cas nuclease. Accordingly, the newly formed guide RNA locus or construct is referred to as a functional guide RNA locus or construct, and the guide RNA encoded by such locus is referred to herein as a functional guide RNA. When the dummy guide RNA constructs are converted to functional guide RNA constructs, such functional guide RNA constructs are more likely to be different from each other, particularly if the library of DNA fragments introduced into the host cell is complex (i.e., has low redundancy).

As will be understood, the number of integrated DNA sequences that each encode a dummy guide RNA is controlled by the end user. In turn, in these embodiments, the cell can only have as many functional guide RNAs as it has integrated DNA sequences that encode the dummy guide RNAs.

All other CRISPR library screens to date have relied upon molecular cloning of the variable/spacer/guide sequences, typically into lentiviral vectors, to achieve libraries of thousands of distinct guide RNAs. This process has proven difficult and laborious. The methods provided herein eliminate the molecular cloning step. Importantly, these methods intrinsically control the number of functional guide RNAs encoded in a cell. The Examples demonstrate the use of these methods in a high-throughput CRISPR/Cas-based library screen.

The foregoing methods can be used to screen for the effect of mutations anywhere in the genome provided a suitable readout is apparent. The mutations that can be studied include mismatch mutations of single or multiple nucleotides such as but not limited to single nucleotide polymorphisms as well as deletions of one or more nucleotides including deletions of genomic regions spanning tens, hundreds, thousands, or more nucleotides. The former can arise if a single cleavage event occurs at a locus. The latter can arise if two relatively proximal guide RNAs are used together to cleave relatively proximal sequences, with the end result being that genomic sequence between the cleavage sites is deleted from the genome.

In some instances, the method further comprises analyzing the effect of the mutation(s) in a defined region. Such regions include for example a regulatory region in the vicinity of the coding region of interest. The expression of the gene (coding sequence) of interest or the expression or functional activity of its gene product (encoded protein) may be used as the readout of the effect of the mutation. Alternatively, a coding sequence for a reporter gene may be inserted into the genome (e.g., in place of the native coding sequence) and its expression or the functional activity of its gene product may be used as the readout. In some instances, the coding sequence of a reporter gene is fused to the native coding sequence, and the readout is the mRNA or protein expression of the resultant fusion protein or the functional activity of the fusion protein. The method can be used to screen and identify sequences involved in cellular processes other than transcription, including for example cell division, cell metabolism, etc. The method can be used to identify mutations that result in loss of function or gain of function, or decrease or increase of transcription. The method may be used to identify the effect of one or more mutations simultaneously. The method may be used to identify the effect of mutations in two or more genes, including two or more regulatory regions, two or more coding sequences, or some combination thereof.

As an example, a population of cells is transfected with a library of DNA fragments each encoding the variable sequence of a guide RNA along with sufficient flanking DNA to enable homologous recombination of the fragment at the dummy guide RNA locus, the DNA fragments are integrated into the dummy guide RNA loci, the functional guide RNAs are expressed in the cells, and in the presence of CRISPR/Cas the functional guide RNAs induce DNA mutations in the cell. The RNA profile and transcriptome of each cell may be analyzed using techniques such as but not limited to single-cell RNA-seq technology. The analysis will reveal the consequence(s) of the genomic mutation on the RNA profile of the cell including the type and abundance of RNA molecules. The method can also be used to identify the nature (e.g., sequence) of the guide RNA that effected the DNA mutation in the first instance. Thus, the effect of a mutation can be observed on the entire cellular transcriptome at once by performing the experiment in a single cell.

In one particular aspect, also provided herein is a method for identifying and elucidating the function of non-coding genomic DNA regions. This method typically inserts a coding sequence for a reporter protein at a genomic locus of interest, and uses the reporter protein as the readout. The resultant locus comprises the coding sequence for a reporter protein (such as for example a fluorescent reporter protein) under the control of regulatory elements of the genomic locus of interest. A CRISPR/Cas system-based library screen is performed in which the genomic region surrounding the reporter coding sequence is “blanketed” with functional guide RNAs which induce a single mutation in each cell at one position. Regulatory regions of interest are then identified as those which result in loss of signal from the reporter protein when mutated. This method takes advantage of the random nature of CRISPR/Cas system mutation. Once such regions of interest are identified, they may be further analyzed using deep sequencing in order to identify the mutation(s) at nucleotide resolution. Analysis of a plurality of cells having altered reporter function (e.g., fluorescence) compared to control (including reduced, absent or enhanced reporter function (e.g., fluorescence)) can be further performed to identify genomic regions that are important in regulation of the reporter coding sequence and thus the native locus of interest. This approach can be used to further our understanding of non-coding DNA regions such as GWAS-identified non-coding SNPs.

Thus, provided herein is a defective CRISPR/Cas “guide RNA” construct comprising a guide sequence that is not homologous to a eukaryotic genome, and a guide hairpin sequence, operably linked to a promoter. In some embodiments, the guide sequence has less than 100% homology to a eukaryotic genome. In some embodiments, the guide sequence has less than 90%, less than 80%, less than 70%, less than 60%, less than 50%, less than 40%, less than 30%, less than 20%, less than 10%, or less than 5% homology to the eukaryotic genome. The guide sequence is sufficiently non-homologous to not trigger a CRISPR/Cas-mediated cleavage event in the eukaryotic genome.

The eukaryotic genome may be a human genome, and thus the defective CRISPR/Cas guide construct may be intended for use in human cells.

The guide sequence may be 19-21 nucleotides in length. The hairpin sequence may be less than 100 nucleotides, less than 80 nucleotides, less than 60 nucleotides, or about 40 nucleotides in length. In other embodiments, the hairpin sequence may be about 20-60 nucleotides in length. Once transcribed, the hairpin sequence can be bound to a CRISPR/Cas nuclease.

The defective CRISPR/Cas guide construct is DNA in nature and when transcribed produces a defective guide RNA.

The construct may be isolated, and thus provided as a single nucleic acid molecule or it may be integrated into the genome of a host cell (i.e., a host cell genome).

Thus, also provided herein is a host cell having integrated into its genome one or more of any of the preceding defective CRISPR/Cas guide constructs. In some embodiments, a single defective CRISPR/Cas guide construct is integrated into the host cell genome. In some embodiments, 2-10 defective CRISPR/Cas guide constructs are integrated into the host cell genome.

Also provided is a population of cells comprising any of the preceding host cells. The population of host cells may be homogeneous or heterogeneous.

Thus, also provided is a cell comprising a CRISPR/Cas dummy guide RNA construct, integrated into the cell genome and comprising a guide sequence that is not homologous to native genomic sequence in the cell.

In some embodiments, the cell comprises a single CRISPR/Cas dummy guide RNA construct. In some embodiments, the cell comprises a coding sequence for a reporter protein or a fusion protein comprising a reporter protein, integrated into the genome.

In some embodiments, the cell further comprises a CRISPR/Cas nuclease and/or a coding sequence for the CRISPR/Cas nuclease. In some embodiments, the cell further comprises a Cas9 nuclease and/or a coding sequence for Cas9 nuclease.

In some embodiments, the cell further comprises a dummy guide RNA sequence that directs a CRISPR-Cas nuclease to the integrated non-homologous guide sequence.

In some embodiments, the cell further comprises a plurality of exogenous DNA fragments, each comprising nucleotide sequences homologous to a native genomic sequence in the cell and to the integrated CRISPR/Cas dummy guide RNA construct. In some embodiments, the exogenous DNA fragments comprise nucleotide sequence from a regulatory region upstream or downstream of a coding sequence of interest. The coding sequence of interest may be coding sequence for a reporter protein or a fusion protein comprising a reporter protein.

Also provided is a cell comprising a CRISPR/Cas functional guide RNA construct, integrated into the cell genome at a defined location, wherein the functional guide RNA construct comprises a guide sequence that is homologous to native genomic sequence in the cell. In some embodiments, the cell comprises more than one CRISPR/Cas functional guide RNA constructs, each integrated into the cell genome at a defined location. In some embodiments, the more than one CRISPR/Cas guide RNA constructs within a single cell comprise different variable guide sequences.

Also provided is a host cell having integrated into its genome a single functional guide RNA construct, in a virus-independent manner.

Also provided is a host cell comprising a plurality of linear, amplified non-integrated DNA fragments, each fragment comprising a CRISPR/Cas variable guide sequence that is homologous to an endogenous (or native) genomic sequence in the host cell and an invariant hairpin sequence that when transcribed is capable of binding a CRISPR/Cas nuclease, wherein the amplified non-integrated DNA fragments are not in a viral vector.

In some embodiments, the host cell has integrated into its genome a coding sequence for a reporter protein or a fusion protein comprising a reporter protein.

Also provided is a method comprising (1) introducing, into a host cell comprising a dummy guide RNA construct, a pool of linear, amplified DNA fragments each comprising a CRISPR/Cas variable guide sequence homologous to a genomic sequence in the cell, and (2) allowing a single linear, amplified DNA fragments to homologously recombine with the dummy guide RNA construct in the presence of a CRISPR/Cas nuclease to form a genomically integrated functional guide RNA construct, wherein the host cell comprises a reporter protein coding sequence integrated into its genome and the linear, amplified DNA fragments are derived from the regulatory region of the reporter protein coding sequence.

In some embodiments, the method further comprises expressing the functional guide RNA that targets a genomic sequence in the host cell, introducing a mutation in the targeted genomic sequence in the presence of a CRISPR/Cas nuclease, and determining the effect of the mutation on the expression of a reporter protein coding sequence.

In some embodiments, the reporter protein coding sequence is a coding sequence of a fusion protein comprising a reporter protein.

In some embodiments, the effect of the mutation is an increase or a decrease in the expression of the reporter protein coding sequence.

In some embodiments, the host cell is in a host cell population and each host cell independently forms a unique functional guide RNA construct.

In some embodiments, each host cell expresses a unique functional guide RNA and is mutated in a different genomic sequence relative to other host cells in the population.

These and other aspects and embodiments of the invention will be described in greater detail herein.

BRIEF DESCRIPTION OF DRAWINGS

FIGS. 1A-1D. Multiplexed Editing Regulatory Assay (MERA). FIG. 1A. In MERA, a genomically integrated dummy gRNA is replaced with a pooled library of gRNAs through CRISPR/Cas9-based homologous recombination such that each cell receives a single gRNA. gRNAs are tiled across the cis-regulatory regions of a GFP-tagged gene locus, and cells are flow cytometrically sorted according to their GFP expression levels. Deep sequencing on each population is used to identify gRNAs preferentially associated with partial or complete loss of gene expression. FIG. 1B. Zfp42^GFP mESCs express uniformly strong GFP. After bulk gRNA integration, a subpopulation of cells lose partial or complete GFP expression. These cells are flow cytometrically isolated for deep sequencing. FIGS. 1C and 1D. Bulk reads for gRNAs are highly correlated between replicates of Tdfg1 (FIG. 1C) or Zfp42 (FIG. 1D), indicating consistent and replicable integration rates.

FIGS. 2A-2D. MERA enables systematic identification of required cis-regulatory elements for Tdgf1. FIG. 2A. A genomic view the Tdgf1 proximal regulatory region showing in track order (i) the location of gRNAs that did not result in GFP loss, (i) enriched gRNAs in GFP^neg cells (dark grey), (iii) enriched gRNAs in GFP^medium cells (light grey), (iv) annotated genes, (v) predicted enhancers (light grey=weak, dark grey=strong), (vi) DNase-I hotspot regions, (vii) transcription factor binding density based on ChIP-seq data, (vii) H3K4me3 ChIP-seq data. Several active regulatory elements coincide with dense clusters of overlapping gRNAs. A large number of gRNA significantly enriched in GFP^neg population are also observed in regions devoid of regulatory element features (UREs). FIG. 2B. Individual validation of specific gRNAs detected as enriched in the GFP^neg population in the MERA assay using the self-cloning CRISPR system. The proportion of cells undergoing GFP loss upon incorporation of a particular gRNA divided by the proportion of cells undergoing GFP loss upon incorporation of GFP-targeting positive control gRNA are plotted against the actual genomic location of the gRNA. Negative controls or gRNA showing no reads in either GFP^neg and GFP^medium populations are highlighted. FIG. 2C. Correlation of gRNAs significantly enriched in the GFP^neg population in fixed size bins varying from 100 bp to 1 kb for biological replicates in Tdgf1. FIG. 2D. Fraction of GFP^neg enriched gRNA among the different functional genomic categories surrounding the Tdgf1 gene.

FIGS. 3A-3C. MERA enables systematic identification of required cis-regulatory elements for Zfp42. FIG. 3A. A genomic view the Zfp42 proximal regulatory region showing in track order (i) the location of gRNAs that did not result in GFP loss, (i) enriched gRNAs in GFP^neg cells (dark grey), (iii) enriched gRNAs in GFP^medium cells (light grey), (iv) annotated genes, (v) predicted enhancers (light grey=weak, dark grey=strong), (vi) DNase-I hotspot regions, (vii) transcription factor binding density based on ChIP-seq data, (vii) H3K4me3 ChIP-seq data. Several active regulatory elements coincide with dense clusters of overlapping gRNAs. FIG. 3B. Correlation of gRNAs significantly enriched in the GFP^neg population in fixed size bins varying from 100 bp to 1 kb for biological replicates in Tdgf1. FIG. 3C. Fraction of GFP^neg enriched gRNA among the different functional genomic categories surrounding the Tdgf1 gene.

FIGS. 4A-4D. Functional motif discovery analysis of region-specific mutant genotypes at enhancers reveals required regulatory motifs. FIG. 4A. A schematic of the procedure involved in finding mutations induced by a particular gRNA. FIG. 4B. Two gRNAs at a proximal Tdgf1 enhancer region in the genomic context showing its overlap with DNAse-I hotspot and predicted enhancer regions, and transcription factor binding sites Stat3, Tcfcp2l1 and Sox2. FIG. 4C. ROC curve for 5-fold classification of GFP^neg and GFP^pos genotypes using mutations within −20 to +20 bp of the gRNA along left and right paired end reads as features. FIG. 4D. Motif logo for region mutated by gRNAs with base scores computed as log-ratios of the Hellinger distance of the GFP^neg genotypes at a base to the reference base to the Hellinger distance of the GFP^pos genotypes at a base to the reference base, caused by Tdgf_gRNA_1 and Tdgf_gRNA_2 along the left paired end read.

FIGS. 5A-5D. Functional motif discovery analysis of a URE reveals critical base positions involved in gene regulation. FIG. 5A. Two gRNAs in the genomic context showing their absence of active histone modifications, known transcription factor binding, predicted enhancers or DNAse-I hotspots. FIG. 5B. ROC curve for 5-fold classification of GFP^neg and GFP^pos genotypes using mutations on the right paired end read within −20 to +20 bp of Tdgf_URE_gRNA2. Unweighted classification (dark grey) counts each unique genotype in the test-set only once while weighted classification (light grey) counts each unique genotype in the test-set as many times as the number of reads assigned to it, for calculating sensitivity and specificity. FIG. 5C. Fraction of unique genotypes in GFP^neg (black plot) and GFP^pos (grey plot) populations with mutations at bases along the right paired end read reveals pattern of cleavage around Tdgf_URE_gRNA2. FIG. 5D. Motif logo for the region mutated by Tdgf_URE_gRNA2 along the right paired end read with base scores computed as log-ratios of the hellinger distance of the GFP^neg genotypes at a base to the reference base to the Hellinger distance of the GFP^pos genotypes at a base to the reference base.

FIGS. 6A-6C. Local genotypes at an enhancer and a URE dictate Tdgf1 expression phenotype. FIG. 6A. Tdgf1 MERA screen ratio of GFP^medium/neg/bulk reads for each gRNA at an upstream enhancer (left) and a downstream URE (right) region. FIG. 6B. Flow cytometric measurement of Tdgf1-GFP expression in clonal cell lines following CRISPR-induced deletion of the shaded regions from (FIG. 6A) show loss of GFP (plots on top panel). CRISPR-mediated homology-directed repair (HDR) back to the wildtype genotype induced robust GFP recovery at both loci (plots on bottom panel). FIG. 6C. Tdgf1 RNA expression in wildtype mESCs (left), clonal mESC lines with deletions of the enhancer and URE shaded in (FIG. 6A) (top panel in FIG. 6B), and bulk mESC lines following HDR back to the wildtype genotype (bottom panel in FIG. 6B), all normalized to wildtype expression level.

DETAILED DESCRIPTION

The invention provides an improved CRISPR/Cas based library screening method. The method does not rely on cloning of guide RNAs into host cells, as is common in prior art methods. Rather, the methods provided herein employ a host cell that harbors one or more dummy guide RNA constructs integrated in its genome. These dummy guide RNA constructs serve as “landing pads” for exogenous variable/spacer/guide sequences introduced into the cell. In the presence of a CRISP/Cas nuclease, such as Cas9, and a dummy guide RNA (having homology to the dummy guide RNA construct), the dummy guide RNA construct is cleaved. In the process of repairing the cleaved site, a single exogenous variable/spacer/guide sequence is introduced into the site through homologous recombination.

The dummy guide RNA construct also serves as a gate keeper. It does so because there will be a 1:1 ratio of dummy to functional guide RNA constructs in the cell following the homologous recombination event. Thus, if the cell harbors one dummy guide RNA construct, it will harbor one functional guide RNA construct; if it harbors two dummy guide RNA constructs, it will harbor two functional guide RNA constructs, and so on.

The screening method has been used to analyze regulatory regions within the genome (i.e., the regulatory genome). Its utility however is not so limited and it can be used to study coding as well as non-coding regions of the genome. The readouts may include a reduction (or an elimination) or an enhancement of gene expression. Moreover, by combining the screening methodology with deep sequencing methods, the regulatory genome can be analyzed at single base resolution.

The method can be used to screen a plurality (e.g., thousands) of sequences in the genome, thereby elucidating the function of such sequences. The method may be used to target mutations in a region of interest (e.g., a region of a chromosome, or region around a gene of interest) or it may be used more broadly. If the mutations are targeted to a gene of interest, then the mutation may be studied by its effect on a coding region of interest or by its effect on a reporter gene that is introduced at or near the site of the coding region of interest. The Examples illustrate methods in which the gene of interest is replaced by or fused to a reporter gene in the form of a fluorescent protein. It should be apparent however that the screening method is not so limited and can be used to introduce mutation in virtually any region of a genome and study the effect of such mutation, thereby elucidating the function of that region of the genome.

If the screen is intended to identify regulatory elements of a particular gene of interest, then this may be accomplished, for example, by creating a library of guide sequences that map to and cover the region of interest. If the guide sequences are contiguous or overlapping, then they may be referred to as “tiled” sequences. The screen can identify regulatory regions of a gene of interest through changes in the expression of the gene (or of a reporter gene inserted at that site, or of a fusion protein of the gene of interest and the reporter gene) as a function of the particular guide RNA harbored and expressed by the cell.

CRISPR/Cas9 Mechanism of Action and Library Screening Rationale

The screening method utilizes the CRISPR/Cas system. Cas9 is a nuclease from the microbial type II CRISPR (clustered regularly interspaced short palindromic repeats) system, which has been shown to cleave DNA when paired with a single-guide RNA (gRNA)^13-16. The gRNA contains a 17-21 bp sequence that directs Cas9 to complementary regions in the genome, thus enabling site-specific creation of double-strand breaks (DSBs) that are repaired in an error-prone fashion by cellular non-homologous end joining (NHEJ) machinery^{14, 15, 17}. Cas9 primarily cleaves genomic sites at which the gRNA sequence is followed by a PAM sequence (−NGG)^18-20, which means that on average ⅛ of all genomic bases can be targeted for Cas9 cleavage, although recent reports have identified Cas9 variants that target distinct PAM sequences which would expand the number of available sites for targeting²¹. NHEJ-mediated repair of Cas9-induced DSBs induces a wide range of mutations initiated at the cleavage site which are typically small (<10 bp) insertion/deletions (indels) but can include larger (>100 bp) indels^{14, 15, 22} and altered individual bases.

Multiplexed Editing Regulatory Assay (MERA)

The screening method described herein may be extended further. An example of such a further extension of this method, referred to as Multiplexed Editing Regulatory Assay (MERA), is described below and in the Examples. The MERA assay has three phases. The first phase is a high-throughput screen that maps necessary genomic elements. This phase is the screening method described above. The second phase functional motif discovery which characterizes regulatory elements identified in the screen. This is accomplished using deep sequencing in and around the region comprising the guide RNA induced mutation in cells or clones having or lacking a phenotype of interest. The third phase is validation of regulatory elements of interest for example by the replacement of selected genomic elements by homologous recombination.

CRISPR/Cas9 has been used in genome-wide mutation screens to identify genes required for survival, drug resistance, and tumor metastasis^23-27 In these screens, gRNAs targeting tens of thousands of sites within genes are cloned into lentiviral vectors and delivered as a pool into target cells along with Cas9. By identifying gRNAs that are enriched or depleted in the cells after selection for the desired phenotype, genes that are required for this phenotype can be systematically identified.

There are at least two distinctions between MERA and previous gene mutation screening approaches. One distinction relates to the observation that the targeted sites are often close together, so cells receiving more than one gRNA may delete a region instead of mutate that region, a phenomenon that would complicate downstream analysis. This issue can be addressed for lentiviral libraries of the prior art by lowering the multiplicity of infection (MOI). The methods provided herein avoid this by limiting the number of guide RNAs in a single cell, typically having only a single guide RNA per cell. Another distinction relates to the fact that a different guide RNA library is required for each gene of interest. All high-throughput CRISPR/Cas9-based approaches to date required cloning of the gRNA libraries into lentiviral vectors and production of library-specific virus. The methods provided herein do not involve cloning and virus production. Instead, the library of the present methods is a library of DNA fragments generated from a region of interest, and optionally amplified and end-modified to facilitate homologous recombination with the dummy RNA locus. The library can be generated and modified more rapidly and used with greater ease as compared to the cloning-based libraries of the prior art.

To enable the efficient targeting of precisely one regulatory element per cell, a strategy was devised to ensure only one (or in some instances a defined or known number of) gRNA can be expressed per cell and that allows gRNA libraries to be used without any molecular cloning into a delivery vector. The Examples demonstrate a system designed with a single copy of a gRNA expression construct integrated into the universally accessible ROSA locus of mESCs using CRISPR/Cas-mediated homologous recombination (FIG. 1A). The construct comprises a U6 promoter driving expression of a dummy guide RNA. CRISPR/Cas-mediated homologous recombination is then used to replace the dummy guide RNA with a functional guide RNA from a library. In the working examples provided herein, an amplification reaction such as PCR was used to add 79-90 bp homology arms to the guide RNA library, since it was found that longer homology arms increase background cutting of unintegrated guide RNA amplified fragments (data not shown) thereby eliminating such unintegrated fragments. The pool of amplified gRNA fragments was then introduced into cells along with a CRISP/Cas system nuclease such as Cas9 and a guide RNA plasmid that induces a DSB in the dummy guide RNA construct. In a substantial fraction of cells (˜30%), the dummy guide RNA construct is repaired by homologous recombination, creating a functional gRNA expression construct targeting a single genomic site from the library (data not shown). It is random which gRNA is integrated in each cell, allowing a pooled screen in which each cell expresses only one gRNA. Of note, the genomic integration-based gRNA screening platform used in MERA could also be applied to other CRISPR-based high-throughput screens as long as the cell line undergoes homologous recombination at appreciable frequency, and it can be modified to achieve expression of any set number of gRNAs per cell for combinatorial screening. While the integration-based approach is thus ill-suited to in vivo screens or screens in cells with limited homologous recombination, it provides an alternative to lentiviral screening that substantially reduces the time, effort, and cost involved in CRISPR-based library screening for applicable cell lines such as ESCs.

Embodiments in which the cell comprises more than one dummy gRNA can be used for a number of applications, including but not limited to combinatorial screening. In some instances, this approach could be used to determine the effect of two or more mutations, either within the same locus or at different genomic loci (e.g., in the same coding sequence, in regulatory regions that control expression of the same coding sequence, etc.), thereby providing insight into how genes or regulatory elements contribute to the phenotype of interest, alone or in combination. As an example, the effect of various combinations of disease-causing SNPs on disease phenotype may be determined. As used herein, a gene refers to coding sequence of a protein and the regulatory regions that control the expression of such coding sequence. The regulatory regions may exist within kilobases or megabases of the coding sequence, and may occur in intronic regions as well.

The host cell may be further modified to express a reporter protein, for example under the control of a regulatory region that is being analyzed using the methods described herein. The coding sequence for the reporter protein can be introduced through a CRISPR-mediated knock-in transgene insertion that uses PCR-amplified homology arms. (Arbab et al. Stem Cell Reports, 2015.) This process avoids the time-consuming step of cloning homology arm vectors as is required in prior art methods. As described in the Examples, this procedure was used to construct GFP knock-in alleles at four genes with mESC-specific expression, Nanog, Rpp25, Tdgf1, and Zfp42. As a proof of principle, mESC-specific genes were chosen because the networks of mESC gene regulation and the epigenetic state of mESC regulatory elements are among the best understood of any cell type^{29, 30}. Two of the genes (Nanog and Rpp25) were constructed as C-terminal GFP fusion proteins, and two (Tdgf1 and Zfp42) had their open reading frames replaced with GFP to compare these two approaches. All knock-in lines, upon clonal selection, had robust GFP expression, although GFP levels were highest in Tdgf1^GFP and Zfp42^GFP lines (FIG. 1B, data not shown). The Nanog^GFP line always contained ˜20% GFP^neg cells, presumably due to known heterogeneous Nanog expression in mESCs (data not shown)³¹.

Four gRNA libraries were constructed, each with 3908 gRNAs tiling cis-regulatory regions of Nanog, Rpp25, Tdgf1, and Zfp42. In the case of Tdgf1, the library targeted the 40 kb region proximal to the gene (coding sequence) in an unbiased manner. In other cases, the regions selected were regions proximal to the gene most likely to be involved in regulation based on enhancer-like features that are a maximum of ˜150 kb away from the coding sequence. Additionally, distal regions up to 92 mB away from the gene were added when ChIA-PET distal interaction data⁹ suggested a possible interaction with the target gene promoter using the Sprout algorithm³⁵. In FIGS. 2A-D and 3A-C, the bulk density panel shows the distribution of integrated guide RNAs along the region probed. Among the 3621 gRNAs found to be integrated in at least 1 replicate of Tdgf1, 99% were within 140 bp of the nearest adjacent gRNA and 95% were within 32 bp of the adjacent gRNA. Distances were measured between the mid-points of adjacent gRNAs. The mean distance between adjacent gRNAs was 11 bp, and the standard deviation of this distance was 32 bp. Of note, repetitive and unmappable genomic regions cannot be tiled with gRNAs, and gRNAs targeting regions whose sequence differs from that of the reference genome cannot be appropriately tiled without genome sequence data of the cell line. Each library also contained 10 positive control gRNAs targeting the GFP open reading frame that we expected would cause GFP loss.

Screening Method Generally

The screening methods described herein utilize a population of genetically modified host cells, and a library (or pool) of linear DNA fragments that comprise a guide sequence homologous to a region of interest in the genome of the host cell. In some instances, the guide sequence is at least 90% or at least 95% homologous to a region of interest. In other instances, the guide sequence is 100% homologous to the region of interest. The starting host cell population is homogeneous: all the cells have one or a controlled and/or defined number of dummy (or defective) guide RNA (gRNA) loci integrated into their genome. As will be understood in the context of this disclosure, a gRNA locus integrated into a cell genome is a DNA that encodes for the gRNA, whether that gRNA is a dummy (or defective) gRNA or a functional gRNA. Cells in a homogenous population will comprise the same gRNA locus or loci. In some instances, all the cells also carry a coding sequence for a reporter protein (reporter gene) or a coding sequence for a fusion of a reporter protein and a protein native to the cell. In addition to their guide sequences, the DNA fragments also comprise sequence homologous to the dummy guide RNA locus. These homologous sequences facilitate homologous recombination of the DNA fragment and the dummy guide RNA locus.

The library of linear DNA fragments is introduced into individual cells in the population. The fragments may be introduced in any number of ways including but not limited to electroporation. Importantly the DNA fragments are not introduced via viral vectors and nor are the DNA fragments provided in the context of viral vectors. Accordingly, the length of such fragments may be 500 nucleotides or less, 400 nucleotides or less, 300 nucleotides or less, 200 nucleotides or less, 100 nucleotides or less, or 50 nucleotides or less in length. The cells then integrate into their genomes one or a controlled (and typically known) number of the fragments. The number of fragments integrated is dictated by the number of dummy guide RNA loci in the genome. Each cell will integrate a fragment independently of other cells.

Also introduced into the cells are a CRISPR/Cas nuclease or a construct that encodes such nuclease and a guide RNA or a construct that encodes such a guide RNA that binds to the dummy guide RNA construct and thereby facilitates cleavage of the construct by the CRISPR/Cas nuclease. The cell is therefore able to produce the nuclease and also to express a guide RNA that targets the dummy guide RNA locus. This results in cleavage of the dummy guide RNA locus by the nuclease, an event that initiates homologous recombination of the dummy guide RNA locus and the DNA fragment via their regions of homology. The resultant locus encodes a functional guide RNA that targets a genomic region, in contrast to the parent dummy guide RNA locus.

The library of DNA fragments is therefore used to generate a library of cells. In some embodiments, each cell is capable of expressing a single CRISPR/Cas guide RNA. In some embodiments, each cell is capable of expressing a controlled number of different CRISPR/Cas guide RNA.

The screening process is then initiated by separating cells that differ in the phenotype of interest, for example by flow cytometric separation of GFP+ and GFP− cells. Alternately, single cell RNA-seq can be performed on all cells without prior selection.

Deletion of Genomic DNA Regions

In another embodiment, defined regions (or blocks) of DNA can be deleted from a genome in each cell and the resulting cells tested for the effect of the deletion as described herein. In this embodiment, a cell expresses two guide RNAs that target genomic locations that define the proximal cut points that flank a block to be deleted. When the guide RNAs are simultaneously expressed, two cuts will be made in the genome, and these cuts will be repaired with the cell's existing machinery for non-homologous end joining. Such blocks can range in size from a few to hundreds of base pairs or more. Block deletion permits more genomic sequence to be tested for function in a single experiment when compared with the embodiment that uses a single guide RNA.

At least two methods can be used to express two guide RNAs in a single cell. A first method uses lentiviral plasmid cloning of paired gRNA libraries that are each expressed off of their own U6 promoter. This method has been described by Vidigal and Ventura (Nature Comm. 2015; 6:8083). These paired guide RNAs can be expressed by lentiviral delivery, or the resulting vector with the guide RNA pair can be PCR amplified with tails that are compatible for homologous recombination into the dummy guide RNA sites that are integrated into the genome as described herein. In the second method, a single RNA construct is expressed from a single U6 promoter and then cleaved by Csy4 into two guide RNAs as described by Tasi et al. (Nature Biotech. 2014; 32(6):569-76). This permits a smaller template to be used that is directly synthesized, and the template can be PCR amplified with tails for homologous recombination into the dummy guide RNA sites that are integrated into the genome as described herein. Alternatively in the second method the paired guide RNAs can be delivered to a cell by lentivirus or other means. Similar approaches may be used to integrate more than two guide RNAs into a single cell.

Guide RNAs

A guide sequence is selected or designed based on the contemplated target sequence. In some embodiments, the target sequence is a sequence within a genome of a cell. Exemplary target sequences include those that are unique in the target genome.

For example, for the S. pyogenes Cas9, a unique target sequence in a genome may include a Cas9 target site of the form M₈N₁₂XGG where N₁₂XGG (N is A, G, T, or C; and X can be anything) has a single occurrence in the genome. A unique target sequence in a genome may include an S. pyogenes Cas9 target site of the form M₉N₁₁XGG where N₁₁XGG (N is A, G, T, or C; and X can be anything) has a single occurrence in the genome.

For the S. thermophilus CRISPR1 Cas9, a unique target sequence in a genome may include a Cas9 target site of the form M₈N₁₂XXAGAAW where N₁₂XXAGAAW (SEQ ID NO: 173) (N is A, G, T, or C; X can be anything; and W is A or T) has a single occurrence in the genome. A unique target sequence in a genome may include an S. thermophilus CRISPR1 Cas9 target site of the form M₉N₁₁XXAGAAW where N₁₁XXAGAAW (SEQ ID NO: 174) (N is A, G, T, or C; X can be anything; and W is A or T) has a single occurrence in the genome.

For the S. pyogenes Cas9, a unique target sequence in a genome may include a Cas9 target site of the form M₈N₁₂XGGXG where N₁₂XGGXG (N is A, G, T, or C; and X can be anything) has a single occurrence in the genome. A unique target sequence in a genome may include an S. pyogenes Cas9 target site of the form M₉N₁₁XGGXG where N₁₁AGGXG (N is A, G, T, or C; and X can be anything) has a single occurrence in the genome. In each of these sequences “M” may be A, G, T, or C, and need not be considered in identifying a sequence as unique.

Further examples of suitable sequences can be found in Chen et al. Cell 2013, 155(7):1479-1491. An example of a suitable sequence is as follows:

This sequence provides a guide sequence of N20 followed by GUUUAAGAGCUAUGCUGGAAACAGCAUAGCAAGUUUAAAU (SEQ ID NO: 2) which as shown above forms one or more hairpins.

It is to be understood that any hairpin sequence can be used provided it can be recognized and bound by a CRISPR/Cas nuclease.

Dummy Guide RNA Constructs

The dummy guide RNA construct is so named because it is unable to target a CRISPR/Cas nuclease to the genome of the host cell, having no variable/guide sequence sufficiently homologous to any region of the host cell genome and thus not capable of effecting a CRISPR//Cas mutation event. Such variable/guide sequence may be referred to herein as being “non-homologous” or defective, intending that it is not able to target a nucleic acid in the host cell using the CRISPR/Cas system. Similarly, the dummy guide RNA construct and the RNA it encodes may be referred to herein as being non-homologous or defective for the same reason.

The dummy guide RNA construct comprises (1) a non-homologous variable/guide sequence and (2) a guide RNA hairpin sequence, and optionally (3) a promoter sequence capable of initiating guide RNA transcription. A non-limiting example of a guide RNA hairpin sequence is the FE hairpin sequence described in Chen et al. Cell. 2013 Dec. 19; 155(7):1479-91. An example of a promoter is the human U6 promoter.

The non-homologous variable/guide sequence may be about 19-21 nucleotides in length. Similarly the variable/guide sequence in each DNA fragment is also about 19-21 nucleotides in length. An example of a non-homologous variable/guide RNA sequence and an invariant guide RNA hairpin sequence are provided in the Examples. It is to be understood that the invention is not so limited and that other non-homologous variant/guide sequences and invariant hairpin sequences may be used.

The dummy guide RNA construct is integrated into the cell at an accessible locus. An example of such a locus is the ROSA26 locus in mouse embryonic stem cells (mESCs). The dummy guide RNA construct may be integrated into the cell through the use of homologous recombination. For example, the construct may comprise sequences that flank the variable/guide sequence and invariant hairpin sequence and that have homology to the desired integration locus.

CRISPR/Cas System Nucleases

In some embodiments, the CRISPR/Cas nuclease is a type II CRISPR/Cas nuclease. In some embodiments, the CRISPR/Cas nuclease is Cas9 nuclease. In some embodiments, the Cas9 nuclease is S. pneumoniae, S. pyogenes, or S. thermophilus Cas9, and may include mutated Cas9 derived from these organisms. The nuclease may be a functionally equivalent variant of Cas9. In some embodiments, the CRISPR/Cas nuclease is codon-optimized for expression in a eukaryotic cell. In some embodiments, the CRISPR/Cas nuclease directs cleavage of one or two strands at the location of the target sequence. The CRISPR/Cas system nucleases include but are not limited to Cas9 and Cpf1.

Host Cells

Virtually any eukaryotic cell type can be used as a host cell provided it can be cultured in vitro and modified as described herein. Preferably, the host cells are an pre-established cell line. The Examples use mESC lines as host cells but the methods of the invention are not so limited. The cells and cell lines may be human cells or cell lines, or they may be non-human, mammalian cells or cell lines.

Amplified DNA Fragments Comprising Guide Sequences

The pool of amplified DNA fragments introduced into the host cell and intended to homologously recombine with the dummy guide RNA construct comprise sequence derived from the host cell genome. In some embodiments, the sequence derived from the host cell genome is intended to act as the guide sequence. When used with CRISPR/Cas nuclease, such guide sequence will target the nuclease to the homologous sequence in the host cell genome.

The guide sequences can be derived from any region of the genome and can be tested for their effect on virtually any function of the cell. Alternatively and as described in greater detail herein, the guide sequence can be derived from a particular region and can be tested for their effect on proximal nucleic acids such as coding sequences including knock-in reporter sequences. In the Examples, the method was used to generate a pool of DNA fragments from the regulatory domains of four transcription factors. The totality of the fragments blanket the regulatory region of interest, whether upstream or downstream of a gene of interest, or both. The totality of the fragments may span+/−10 kb, or +/−20 kb, or +/−50 kb, or +/−100 kb, or more of regulatory sequence (upstream and downstream) of the gene of interest. The variable/guide sequences are typically about 19-20 nucleotides in length.

Reporter Genes and Proteins, and Readouts

The reporter gene may be integrated into the cells using a CRISPR/Cas mechanism, in some embodiments. For example, to generate a “knock-in” of a reporter construct, a similar approach to that used to integrate a DNA fragment into a dummy guide RNA construct locus. For example, an expression vector, such as a plasmid, may be used that comprises a promoter (e.g., U6 promoter), a guide RNA hairpin sequence, and a guide sequence that targets the desired genomic locus where the reporter construct is to be integrated. Such an expression vector may be generated by cloning the guide sequence into an expression construct comprising the remaining elements. A DNA fragment comprising the coding sequence for the reporter protein can be generated and subsequently modified to include homology arms that flank the coding sequence of the reporter protein. The guide RNA expression vector, the amplified DNA fragments comprising the reporter protein coding sequence, and a CRISPR/Cas nuclease (or an expression vector encoding the nuclease) are introduced into the host cell (e.g., via electroporation). The expression vectors may further comprise additionally selection markers such as antibiotic resistance markers to enrich for cells successfully transfected with the expression vectors. Cells that express the reporter protein can be further selected. Exemplary sequences for the guide RNAs, homology arms and genomic integration sites for particular mESC loci are provided in the Examples.

Reporter genes are used for identifying potentially transfected cells and for evaluating the functionality of regulatory sequences. In general, a reporter gene is a gene that is not endogenous or native to the host cells and that encodes a protein that can be readily assayed. Reporter genes that encode for easily assayable proteins are known in the art and are typically preferred.

The reporter proteins may be selected from a variety of reporter systems available in the art. The Examples use green fluorescent protein (GFP) as the reporter protein but the invention is not so limited. Examples of reporter genes include, but are not limited to, glutathione-S-transferase (GST), horseradish peroxidase (HRP), chloramphenicol acetyltransferase (CAT) beta-galactosidase, beta-glucuronidase, luciferase, HcRed, DsRed, cyan fluorescent protein (CFP), yellow fluorescent protein (YFP), and autofluorescent proteins including blue fluorescent protein (BFP), cell surface markers, antibiotic resistance genes such as neo, and the like.

It is to be understood that reporter genes or other selectable marker genes can be used in the methods described herein to identify and/or select cells from the transfected population. In some instances, the selectable marker or reporter genes may be flanked with appropriate regulatory sequences to enable expression in the host cells. In some instances, the selectable marker or reporter gene is flanked by a native regulatory region that will be targeted using the CRISPR/Cas mutation strategy described herein.

Expression Vectors

The term “vector” refers to a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. Vectors include, but are not limited to, nucleic acid molecules that are single-stranded, double-stranded, or partially double-stranded; nucleic acid molecules that comprise one or more free ends, no free ends (e.g. circular); nucleic acid molecules that comprise DNA, RNA, or both; and other varieties of polynucleotides known in the art. One type of vector is a “plasmid,” which refers to a circular double stranded DNA loop into which additional DNA segments can be inserted, such as by standard molecular cloning techniques. Certain vectors are capable of autonomous replication in a host cell into which they are introduced (e.g. bacterial vectors having a bacterial origin of replication and episomal mammalian vectors). Other vectors (e.g., non-episomal mammalian vectors) are integrated into the genome of a host cell upon introduction into the host cell, and thereby are replicated along with the host genome. Moreover, certain vectors are capable of directing the expression of genes to which they are operatively-linked. Such vectors are referred to herein as “expression vectors.” Expression vectors in recombinant DNA techniques often take the form of plasmids.

Recombinant expression vectors can comprise a nucleic acid of the invention in a form suitable for expression of the nucleic acid in a host cell, which means that the recombinant expression vectors include one or more regulatory elements, which may be selected on the basis of the host cells to be used for expression, that is operatively-linked to the nucleic acid sequence to be expressed. Within a recombinant expression vector, “operably linked” is intended to mean that the nucleotide sequence of interest is linked to the regulatory element(s) in a manner that allows for expression of the nucleotide sequence (e.g. in an in vitro transcription/translation system or in a host cell when the vector is introduced into the host cell).

“Complementarity” refers to the ability of a nucleic acid to form hydrogen bond(s) with another nucleic acid sequence by either traditional Watson-Crick or other non-traditional types.

A percent complementarity indicates the percentage of residues in a nucleic acid molecule which can form hydrogen bonds (e.g., Watson-Crick base pairing) with a second nucleic acid sequence (e.g., 5, 6, 7, 8, 9, 10 out of 10 being 50%, 60%, 70%, 80%, 90%, and 100% complementary). “Perfectly complementary” means that all the contiguous residues of a nucleic acid sequence will hydrogen bond with the same number of contiguous residues in a second nucleic acid sequence. “Substantially complementary” as used herein refers to a degree of complementarity that is at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 97%, 98%, 99%, or 100% over a region of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, or more nucleotides, or refers to two nucleic acids that hybridize under stringent conditions.

The term “homologous” as used herein refers to a nucleic acid sequence that has complementarity to another nucleic acid sequence or to the complement of such sequence. Two sequences can have a degree of homology that is at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 97%, 98%, 99%, or 100% over a region of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, or more nucleotides. Two sequences that are homologous are able to hybridize to each other under stringent conditions.

The following Examples are included for purposes of illustration and are not intended to limit the scope of the invention.

EXAMPLES

Introduction

We introduce the Multiplexed Editing Regulatory Assay (MERA), a high-throughput CRISPR/Cas9-based approach that analyzes the regulatory genome for function in its native context. By tiling thousands of mutations across ˜40 kb of cis-regulatory genomic space and using knock-in GFP reporters to read out gene activity, we obtain quantitative information on the contribution of cis-regulatory regions to gene expression. We identify proximal and distal regulatory elements necessary for gene expression of four embryonic stem cell-specific genes. We show a consistent contribution of neighboring gene promoters to gene expression, and we identify a novel class of unmarked regulatory elements (UREs) that control gene expression but do not have typical enhancer epigenetic or chromatin features. Comparing thousands of functional and non-functional genotypes at a genomic location, we find the base pair-resolution functional motifs of regulatory elements.

Experimental Design and Overview

Library Design for MERA Assay

In addition to 10 GFP-targeting gRNAs we designed 3908 gRNAs specific to each of the four libraries for TDGF, Nanog, Zfp42 and Rpp25. For TDGF we selected a −20 kb to +20 kb proximal region around the TDGF promoter to profile 3908 gRNAs that were designed for this region. For Nanog, Rpp25, and Zfp42 we prioritized the design of 3908 gRNAs based on regions of strong DNAse-I enrichment going up to 100 kb on either side of the gene promoter. Further, we used PolII Chia-Pet data to find distal regions that are predicted to interact with the promoter. In case of a large number of Chia-pet regions, we filtered interactions based on other enhancer features such as p300 binding, DNASE-I enrichment, active histone modifications etc. overlapping distal Chia-pet regions.

Libraries were ordered as 98-100 bp sequences containing a 19-20 bp protospacer targeting the genomic sequence of interest (i.e., variable or guide sequence), an optional G if the protospacer does not already begin with one, and surrounding sequences homologous to the U6 promoter and gRNA hairpin. The format was as follows: TTATATATCTTGTGGAAAGGACGAAACACC[GN_18-20]GTTTAAGAGCTATGCTGGAAA CAGCATAGCAAGTTTAAATAAGGCTAGT (SEQ ID NO:3) Libraries were ordered from LC Sciences.

Mapping of MERA Reads

We mapped the sequence composing of sample barcode, primer and exact matches of the designed gRNA sequence to the sequenced reads. Counts for each gRNA for either GFP^neg, GFP^medium or bulk populations were obtained by counting the number of sequenced reads that showed exact matches to the gRNA.

The gRNA integration rate into cellular genomic DNA was found to be 93% for Tdgf1 but appeared to be only 43% for Zfp42. In order to determine if this was caused by inefficient integration or due to synthesis errors, we sequenced the gRNA library for Zfp42 and found that only 1723 of the 3919 guide RNAs in the Zfp42 library were synthesized accurately. Among these, 1718/1723 were detected in the bulk library of at least one replicate. Hence, we estimate that the integration rate of gRNAs is >90% of those that are synthesized. Oligonucleotide library synthesis quality is unaffected by whether a gRNA integration approach such as MERA or a lentiviral cloning approach is taken, and thus MERA enables integration of the vast majority of available gRNAs.

Identification of gRNAs that are Significantly Enriched in GFP^neg and GFP^medium Populations

In order to detect gRNAs with statistically significant overrepresentation in GFP^neg and GFP^medium populations, we perform a step-wise procedure.

Step 1. We normalize the gRNA sequence read counts, which can vary between sequencing runs of bulk, GFP^medium and GFP^neg populations due to differences in cell number and diversity of the respective populations (data not shown). In order to normalize these read ranges, we assume that the positive control gRNAs targeting the GFP coding region always induce loss of GFP expression, which is consistent with our previous results showing that over 99% of cells receiving a GFP-targeting gRNA lose GFP expression (Arbab et al. Stem Cell Reports, 2015). In addition, GFP^neg, and to a lesser extent GFP^med reads are always observed to be proportional to the bulk reads for the =targeting guide RNAs, to a much greater extent than for all guide RNAs (data not shown). Hence, we predict the number of GFP^neg reads we would see for each gRNA given its bulk and GFP^med count if it always caused GFP loss. In order to do this, we build two different kinds of linear models depending on the data available

I. In case of Tdgf1^GFP and Zfp42^GFP, we have a GFP^medium as well as GFP^neg population, along with 3 to 4 biological replicates per cell-line. We assume that for any GFP-targeting gRNA, the majority of bulk reads are derived from the GFP^neg population. However, each gRNA may also cause some intermediate loss of GFP due to variable mutations or imperfect sorting. In addition, there is a low gRNA-dependant intercept or GFP^pos population, which may be a small fraction of mutations induced by a particular gRNA that do not cause GFP-loss.

In order to transform the bulk reads to the GFP^neg scale, we model GFP^neg as the dependant variable, and GFP^medium and bulk reads as independent variables using a generalized linear model (Nelder and Wedderbu, J R Stat Soc Ser a-G 135:370 (1972)). The intercept is modeled as being dependent on the gRNA but independent across replicates, while the slopes are considered as having a replicate-dependent component also.

The model is of the form
y˜x1+x2+(z11|g1)+(x1|g2)+(x2|g2)
where,
y=GFP^neg
x1=Bulk
x2=GFP^medium
z11=Intercept
g1=grouping by gRNA
g2=grouping by replicate

In order to transform the bulk reads to the GFP^medium scale, we use the same model but with

y=GFP^medium

x2=GFP^neg

II. In case of Nanog^GFP and Rpp25^GFP, we have only a GFP^neg population and at most 2 replicates. In this case we build an independent linear regression model for each replicate of the form:
y˜x1+z11,
where
y=GFP^neg
x1=Bulk
z11=Intercept

Using the linear regression models, we now transform all bulk reads to either GFP^neg or GFP^medium populations, depending on if we are interested in finding gRNAs enriched in GFP^neg or GFP^medium populations respectively.

Step 2. We now use the fact that since the dummy gRNA (negative control) should not occur in GFP^neg/GFP^medium cells any reads corresponding to this gRNA in the GFP^neg/GFP^medium population are due to random chance. Hence, we can obtain the null probability of observing reads in the GFP^neg/GFP^medium population by dividing the GFP^neg/GFP^medium reads for the dummy gRNA by the number of bulk reads for the dummy gRNA transformed to the GFP^neg/GFP^medium scale. We then use a binomial distribution to calculate significance for a gRNA based on this null probability, with the gRNA's observed number of GFP^neg/GFP^medium reads as the number of successes, and the number of bulk-transformed reads for the gRNA as the number of trials.

Datasets for Comparison and Visualization with Enriched gRNA

The UCSC genome browser (Kent et al. Genome research 12: 996-1006 (2002)) was used to visualize the data and create genomic view snapshots for regulatory regions of various genes.

Enhancer Predictions

The enhancer predictions were made using the RFECS method (Rajagopal et al. PLoS computational biology 9: e1002968 (2013)) using 6 histone modifications from ENCODE (Nature 489:57-74 (2012)) trained on p300 binding site data from mouse embryonic stem cells. Enhancers were separated into “strong” and “weak” categories based on presence of H3K27ac at levels greater than input. Further boundaries of enhancers were called using a Sobel edge-detection algorithm implemented in MATLAB. Edges were identified for an input subtracted RPKM (reads per kilobase per million)-normalized H3K27ac reads (Rajagopal et al. PLoS computational biology 9: e1002968 (2013)) in the case of strong enhancers and RPKM-normalized H3K4me1 reads for weak enhancers.

DNase-I hotspot: We used the DNase-seq dataset previously generated (Sherwood et al. Nature biotechnology 32: 171-178 (2014)) and called hotspots using a standard hotspot algorithm (John et al. Nature genetics 43:264-268 (2011)).

TF density: The GEM algorithm (Guo et al. PLoS computational biology 8: e1002638 (2012)) was applied to transcription factor Chip-seq datasets for the following transcription factors: Nanog, Oct4, Sox2, TCF3, p300, CTCF, Smc1, Smad3, c-Myc, Med12, Med1, CTCF, E2F1, Esrrb, Klf4, n-Myc, Nr5a2, Tcfcp211, Stat3, Zfx.

Analysis of Deep Sequencing Datasets

Individual scCRISPR-mediated mutation by a selected gRNA was performed in a large pool of cells to create tens of thousands of unique mutated genotypes at the site. We then flow cytometrically sorted GFP^pos and GFP^medium/neg populations and performed 150 bp paired-end sequencing on regions surrounding each targeted site to obtain genotypic data on thousands of mutated regions that did and did not induce loss of GFP expression (FIG. 4A). Deep-sequencing datasets were filtered for sequence quality by using a minimum base quality filter of 30. After stripping barcodes, the length of each paired end read was 145 bp. We aligned these 145 bp long genotypes to the reference genotype extended by 30 bp downstream (total of 175 bp). Alignment of sequenced reads to the reference genome was performed using the semi-global version of the Needleman Wunsch algorithm with a gap opening penalty of 8 and gap extension penalty of 4. The command in MATLAB used was:

nwalign(Reference_Seq,

Genotype_seq,‘alphabet’,‘NT’,‘gapopen’,8,‘ExtendGap’,4,‘glocal’,‘true’);

Functional Motif Discovery

After globally aligning and filtering reads for sequence quality (per base quality >=30), mismatches, deletions and insertions were counted with respect to the base position in the reference. We observed long stretches of mutations with combinations of mismatches and deletions. Hence, we defined a “length of disruption” as a continuous series of mutations with maximum intervening matches of <5 bases. We plotted the left and right ends of these disruptions and observed the majority of disruptions originated within the gRNA as expected with very few short mutations lying outside that could be assumed to be one or two base sequencing errors (data not shown). While a majority of disruptions extending beyond the ends of the guide RNA were enriched for GFP^neg (data not shown), we also observed a mixed population of GFP^neg as well as GFP^pos deletions lying within −20/+20 bp of the gRNA. Since we wish to assess the local effect of the gRNA on GFP-loss, we limited further analysis to genotypes with disruptions that originate within the gRNA and do not go beyond 20 bp of the gRNA.

Restricting our analysis to these genotypes, we observed increased mutation around the gRNA cleavage site in both GFP^pos and GFP^medium/neg populations (data not shown). Mismatch, deletion, and insertion mutations were all observed, with deletions predominating in the GFP^medium/neg genotypes (data not shown).

In order to develop a base-level motif logo, we defined a base-level score representing the deviation of GFP^neg population from reference as compared to the deviation of the GFP^pos population from reference. In order to find the distance of a base from reference, we used the Hellinger measure (Liese and Miescke, Statistical Decision Theory: Estimation, Testing, and Selection. Springer Ser Stat, 1-677 (2008)) for finding the distance between two discrete distributions:

$H\left(P,Q\right)=\frac{1}{\sqrt{2}}\sqrt{\sum _{i=1}^k{\left(\sqrt{p_i}-\sqrt{q_i}\right)}^2},$

Here, we had five possible values per base which were the frequency of occurrence of each base type (A,C,T,G) and a fifth deletion (D). The motif score at any base was defined as:

Base score=log 10 (H(GFP^neg, Reference)/H(GFP^pos, Reference)) These base scores were plotted as a motif logo along −20/+20 bp of the gRNA to indicate relative importance of each base, independent of the cutting biases of the gRNA. It should be noted that since all mutations for GFP^pos as well as GFP^neg arise within the seed region of the gRNA, it is sometimes difficult to obtain a base-level importance score for these bases surrounding the cleavage site. However, due to the random lengths of stretches of mutations originating from the cleavage site we can observe distinct sequence profiles emerging upstream and downstream of these bases.

Classification of GFP^pos and GFP^neg Populations

We represented mismatches, insertions and deletions within −20/+20 bp of the gRNA as features. For all of the bases within the gRNA we represented 5 possibilities—A, C, T, G, and deletion. The feature for a base was one of four values for a particular base or the integer number of deleted bases starting at that base. Converting this categorical representation to a numeric format, we obtained 5×(length of gRNA+40) features. Insertions were represented as the integer number of bases inserted immediately after each base of the gRNA and flanking boundaries. Hence, the total features were=6×(length of gRNA+40)

We performed 5-fold classification of unique genotypes in GFP^pos and GFP^neg populations using a parallelized random forest implemented in MATLAB. We used 100 trees and ascertained that the out-of-bag classification error had reached convergence at this parameter value. Classification rate for a test-set genotype, was computed in an unweighted manner by counting each test-set genotype only once. In case of weighted accuracy measures, we weighted the accuracy of classification for each test-set genotype, by the number of reads assigned to it.

Conservation of Bases

We examined the vertebrate phastcons score for every base in the gRNA at the URE to see if there was a correspondence with the importance of the base for regulation as determined above (data not shown).

Materials and Methods

Cell Culture

Mouse embryonic stem cell (mESC) culture was performed according to previously published protocols (Nature 489:57-74 (2012)). All experiments were performed with 129P2/OlaHsd mESCs. mESCs were maintained on gelatin-coated plates feeder-free in mESC media composed of Knockout DMEM (Life Technologies) supplemented with 15% defined fetal bovine serum (FBS) (HyClone), 0.1 mM nonessential amino acids (NEAA) (Life Technologies), Glutamax (GM) (Life Technologies), 0.55 mM 2-mercaptoethanol (b-ME) (Sigma), 1×ESGRO LIF (Millipore), 5 nM GSK-3 inhibitor XV and 500 nM U0126. Cells were regularly tested for mycoplasma.

ROSA26 gRNA Cassette Knock-In

Our first step was to derive a mESC line with a dummy gRNA in the universally accessible ROSA26 locus (Sherwood et al. Nature biotechnology 32: 171-178 (2014)). We did so using the following protocol, which involves amplification of the desired dummy gRNA knock-in construct with short PCR-amplified homology arms using a protocol we have recently developed (Arbab et al. Stem Cell Reports, 2015):

1. PCR amplify homology arm-flanked dummy gRNA in two successive PCR steps.

PCR a plasmid containing the dummy gRNA downstream of the U6 promoter and including the “FE” modified gRNA hairpin (John et al. Nature genetics 43:264-268 (2011)) with primers that amplify the entire expression cassette with ROSA26 homology arms:

1. 091514_U6gRNA_ROSAHDR_fw

(SEQ ID NO: 4)

CCAGGTTAGCCTTTAAGCCTGCCCAGAAGA CTCCCGCCCA GCATGTGA

GGGCCTATTTCC

2. 091514_U6gRNA_ROSAHDR_rv

(SEQ ID NO: 5)

GGAGAATCCCTTCCCCCTCTTCCCTCGTGAT CTGCA TCGCGATTTTAC

CACATTTGTAGA

Use 2× Phusion PCR Mastermix (NEB), 25 uL reaction, 35 cycles of 2-step PCR (98 for 10s, 72 for 45s) using 3% DMSO

Test ˜2 uL, expecting 750 bp band.

Use 10 uL unpurified product in a new reaction with the following primers that extend the ROSA26 homology arm to enable high-efficiency genomic knock-in of the dummy gRNA expression cassette:

3. 091514_ROSAHDR_Ext_fw

(SEQ ID NO: 6)

ACACCTGTTCAATTCCCCTGCAGGACAACGCCCA CACACCAGGTTAGCC

TTTAAGCCTGC

4. 091514_ROSAHDR_40bpext_rv

(SEQ ID NO: 7)

TCTGCTGCCTCCTGGCTTCTGAGGACCGCCCT GGGCCTGGGAGAATCCC

TTCCCCCTCTT

Use a 200 uL Phusion reaction, 35 cycles of 2-step PCR (98 for 10s, 72 for 45s) using 3% DMSO

Test ˜2 uL, expecting 800 bp band.

Minelute PCR purify the product in 10 uL elution buffer.

2. Co-electroporate mESC with Cas9, gRNA, and PCR-amplified homology arm construct to generate ROSA26 locus dummy gRNA

a. Co-electroporate 1 well of a 6-well plate of mESC with:

5 ug p2T CBh Cas9 BlastR

5 ug p2T U6sgROSA26-FE HygroR (p2T sg82 HygR)

MinElute-purified ROSA-HDR U6sgIhhPro-FE from 200 uL PCR reaction

Vacuum concentrate to <20 uL volume of DNA, use 120 uL EmbryoMax Electroporation Buffer (ES-003-D, Millipore). DNA mixture and mESC suspension were mixed and electroporated in a 0.4 cm electroporation cuvette using a Bio-Rad electroporator at 230 V, 0.500 uF, and maximum resistance. Transiently select with Blasticidin and Hygromycin from 24-72 hours after electroporation (total of 48 hours of selection).

We identified clonal knock-in lines through mESC colony genomic DNA PCR with the following primers, identifying a 134-bp knock-in band:

5. 080814_U6gRNA_late_fw

(SEQ ID NO: 8)

	TCTACAAATGTGGTAAAATCGCGA
	091514_ROSA_downstream_rv

(SEQ ID NO: 9)

GGGAGGGGAGTGTTGCAATA

We sequence verified the knock-in lines using the following primers, isolating and sequencing the 913 bp knock-in band:

6. 091514_ROSA_upstream_fw

(SEQ ID NO: 10)

	TGGGAAGTCTTGTCCCTCCA
	7. 091514_ROSA_downstream_rv

(SEQ ID NO: 9)

GGGAGGGGAGTGTTGCAATA

We thus obtained a heterozygous knock-in cell line which guarantees that only one gRNA construct can be integrated and expressed per cell in the context of pooled screening. Sequences of the ROSA26 locus are below:

Wildtype ROSA26 locus

(SEQ ID NO: 11)

TCCCATTTTCCTTATTTGCCCCTATTAAAAAACTTCCCGACAAAACCGAA

AATCTGTGGGAAGTCTTGTCCCTCCAATTTTACACCTGTTCAATTCCCCT

GCAGGACAACGCCCACACACCAGGTTAGCCTTTAAGCCTGCCCAGAAGAC

TCCCGCCCATCTTCTAGAAAGACTGGAGTTGCAGATCACGAGGGAAGAGG

GGGAAGGGATTCTCCCAGGCCCAGGGCGGTCCTCAGAAGCCAGGAGGCAG

CAGAGAACTCCCAGAAAGGTATTGCAACACTCCCCTCCCCCCTCCGGAGA

AGGGTGCGGCCTTCTCCCCGCCTACTCCAC

ROSA26 with U6 sgdummygRNA-FE knock-in

(SEQ ID NO: 12)

TCCCATTTTCCTTATTTGCCCCTATTAAAAAACTTCCCGACAAAACCGAA

AATCTGTGGGAAGTCTTGTCCCTCCAATTTTACACCTGTTCAATTCCCCT

GCAGGACAACGCCCACACACCAGGTTAGCCTTTAAGCCTGCCCAGAAGAC

TCCCGCCCAGCATGTGAGGGCCTATTTCCCATGATTCCTTCATATTTGCA

TATACGATACAAGGCTGTTAGAGAGATAATTGGAATTAATTTGACTGTAA

ACACAAAGATATTAGTACAAAATACGTGACGTAGAAAGTAATAATTTCTT

GGGTAGTTTGCAGTTTTAAAATTATGTTTTAAAATGGACTATCATATGCT

TACCGTAACTTGAAAGTATTTCGATTTCTTGGCTTTATATATCTTGTGGA

AAGGACGAAACACCGAGGCGTCTGGGTGGCTCTTGGTTTAAGAGCTATGC

TGGAAACAGCATAGCAAGTTTAAATAAGGCTAGTCCGTTATCAACTTGAA

AAAGTGGCACCGAGTCGGTGCTTTTTTGTTTTAGAGCTAGAAATAGCAAG

TTAAAATAAGGCTAGTCCGTTTTTAGCGCGTGCGCCAATTCTGCAGACAA

ATGGCTCTAGAGGTACGGCCGCTTCGAGCAGACATGATAAGATACATTGA

TGAGTTTGGACAAACCACAACTAGAATGCAGTGAAAAAAATGCTTTATTT

GTGAAATTTGTGATGCTATTGCTTTATTTGTAACCATTATAAGCTGCAAT

AAACAAGTTAACAACAACAATTGCATTCATTTTATGTTTCAGGTTCAGGG

GGAGATGTGGGAGGTTTTTTAAAGCAAGTAAAACCTCTACAAATGTGGTA

AAATCGCGATGCAGATCACGAGGGAAGAGGGGGAAGGGATTCTCCCAGGC

CCAGGGCGGTCCTCAGAAGCCAGGAGGCAGCAGAGAACTCCCAGAAAGGT

ATTGCAACACTCCCCTCCCCCCTCCGGAGAAGGGTGCGGCCTTCTCCCCG

CCTACTCCAC

We then proceeded to gRNA screening. We use homologous recombination inside target cells to replace the dummy gRNA with a pool of amplified gRNAs that contain homology tails matching the dummy gRNA and thus will seamlessly integrate as functional gRNAs. To do so, we co-electroporated Cas9, a gRNA plasmid that cuts the dummy gRNA, and a library of replacement gRNA PCR fragments. The sequence of the gRNA that cuts the dummy gRNA is

(SEQ ID NO: 13)

GAAACACCGAGGCGTCTGGG

GFP Fusion/Knock-in Line Generation

We generated gene knock-in constructs with short PCR-amplified homology arms using a protocol we have recently developed (Arbab et al. Stem Cell Reports, 2015). For each knock-in, we used mESCs with an integrated ROSA26 gRNA cassette, constructed as described above.

For Zfp42 and Tdgf1, we constructed GFP replacement alleles in which we added GFP at the translational start site of the genes. This approach should disrupt endogenous gene activity in a single allele of that gene. For Nanog and Rpp25, we constructed C-terminal GFP fusion proteins in which the full-length protein is fused to GFP. This approach should maintain protein functionality.

To perform knock-in, an gRNA targeting the desired genomic insertion site was cloned into a plasmid containing a U6 promoter, gRNA hairpin, and Hygromycin resistance cassette. GFP was amplified in two successive steps with homology arm primers adding 70-80 bp of homologous sequence surrounding the desired insertion site to GFP. Then mESCs were co-electroporated with the gRNA plasmid, the GFP homology arm amplicon, and a plasmid expressing Cas9 and a Blasticidin resistance cassette. Transient Blasticidin and Hygromycin selection was performed from 24-72 hours after electroporation, then GFP-expressing mESCs were flow cytometrically sorted and then clonally purified. Genomic DNA PCRs confirmed site-specific genomic integration of GFP. An expanded protocol for this procedure is contained in Arbab et al. Stem Cell Reports, 2015, and the sequences of gRNAs, homology arms, and genomic integration sites are shown below.

Tdgf1

Tdgf1 gRNA:

(SEQ ID NO: 14)

GAGATGGGGTACTTCTCATCC

Tdgf1 genomic region:

(SEQ ID NO: 15)

TTGGGTGTTTCGAGAATGGCTTTATGAACTAAAGCCATCTGCTAATATTG

TGTTTCTTGTCTTTTCCTCCAACGTTTTTACGAGCCGTCGAAGATGGGGT

ACTTCTCATCCAGGTATGAGCTAACCTTGACTTTTTGGTTGCTGGAGATA

GCCACTTCGGAAAATCACGTTCTATGACGCTCTGATTTTTGTCTTGCTTG

AACCTTGTCAGTAAC

GFP in Tdgf1 locus

(SEQ ID NO: 16)

CTCTCTCATTTGGCATATCTTTCTTTTTAATCTACTGTTTTCATTTTGTG

AAATTAGCCTTTGGGTGTTTCGAGAATGGCTTTATGAACTAAAGCCATCT

GCTAATATTGTGTTTCTTGTCTTTTCCTCCAACGTTTTTACGAGCCGTCG

AAGATGGTGAGCAAGGGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCT

GGTCGAGCTGGACGGCGACGTAAACGGCCACAAGTTCAGCGTGTCCGGCG

AGGGCGAGGGCGATGCCACCTACGGCAAGCTGACCCTGAAGTTCATCTGC

ACCACCGGCAAGCTGCCCGTGCCCTGGCCCACCCTCGTGACCACCCTGAC

CTACGGCGTGCAGTGCTTCAGCCGCTACCCCGACCACATGAAGCAGCACG

ACTTCTTCAAGTCCGCCATGCCCGAAGGCTACGTCCAGGAGCGCACCATC

TTCTTCAAGGACGACGGCAACTACAAGACCCGCGCCGAGGTGAAGTTCGA

GGGCGACACCCTGGTGAACCGCATCGAGCTGAAGGGCATCGACTTCAAGG

AGGACGGCAACATCCTGGGGCACAAGCTGGAGTACAACTACAACAGCCAC

AACGTCTATATCATGGCCGACAAGCAGAAGAACGGCATCAAGGTGAACTT

CAAGATCCGCCACAACATCGAGGACGGCAGCGTGCAGCTCGCCGACCACT

ACCAGCAGAACACCCCCATCGGCGACGGCCCCGTGCTGCTGCCCGACAAC

CACTACCTGAGCACCCAGTCCGCCCTGAGCAAAGACCCCAACGAGAAGCG

CGATCACATGGTCCTGCTGGAGTTCGTGACCGCCGCCGGGATCACTCTCG

GCATGGACGAGCTGTACAAGTAAAGCGGCCGCAATTCACTCCTCAGGTGC

AGGCTGCCTATCAGAAGGTGGTGGCTGGTGTGGCCAATGCCCTTGCTCAC

CATGGTGAAGGGTGGGCGCGCCGACCCAGCTTTCTTGTACAAAGTGGTTG

ATCTAGAGGGCCCGCGGTTCGAAGGTAAGCCTATCCCTAACCCTCTCCTC

GGTCTCGATTCTACGCGTACCGGTCATCATCACCATCACCATTGAGTTTA

TCCATCACACTGGCGGCCGCTCGAGGGGGATCCACTAGTTCTAGAGCGGC

CGCTTCGAGCAGACATGATAAGATACATTGATGAGTTTGGACAAACCACA

ACTAGAATGCAGTGAAAAAAATGCTTTATTTGTGAAATTTGTGATGCTAT

TGCTTTATTTGTAACCATTATAAGCTGCAATAAACAAGTTAACAACAACA

ATTGCATTCATTTTATGTTTCAGGTTCAGGGGGAGATGTGGGAGGTTTTT

TAAAGCAAGTAAAACCTCTACAAATGTGGTAAAATCGCGAGAGCTAACCT

TGACTTTTTGGTTGCTGGAGATAGCCACTTCGGAAAATCACGTTCTATGA

CGCTCTGATTTTTGTCTTGCTTGAACCTTGTCAGTAACATTGCTGCTTTT

CCTGAAGAACCTGGAACTT

Zfp42

Zfp42 gRNA:

(SEQ ID NO: 17)

GAATGAACAAATGAAGAAAA

Zfp42 genomic region:

(SEQ ID NO: 18)

TGGGTTATTATCTAAGGCAGGTGTTTGCGGATCAGTGCCCCCTGGAAGTG

AGTCATAGGCATTGTTCAAGAAGGAAGCAGCTAAGACAACATGAATGAAC

AAAAAATGAATGAACAAATGAAGAAAACGGCAAAGACAAGTGGCCAGAAA

GGGCCGGGCGGAAGAGCCCTCGACAGACTGACCCTAAAGCAAGACGAGGC

AAGGCCAGTCCAGAATACCAGAGTGG

GFP in Zfp42 locus:

(SEQ ID NO: 19)

TGGGTTATTATCTAAGGCAGGTGTTTGCGGATCAGTGCCCCCTGGAAGTG

AGTCATAGGCATTGTTCAAGAAGGAAGCAGCTAAGACAACATGAATGAAC

AAAAAATGAATGTGAGCAAGGGCGAGGAGCTGTTCACCGGGGTGGTGCCC

ATCCTGGTCGAGCTGGACGGCGACGTAAACGGCCACAAGTTCAGCGTGTC

CGGCGAGGGCGAGGGCGATGCCACCTACGGCAAGCTGACCCTGAAGTTCA

TCTGCACCACCGGCAAGCTGCCCGTGCCCTGGCCCACCCTCGTGACCACC

CTGACCTACGGCGTGCAGTGCTTCAGCCGCTACCCCGACCACATGAAGCA

GCACGACTTCTTCAAGTCCGCCATGCCCGAAGGCTACGTCCAGGAGCGCA

CCATCTTCTTCAAGGACGACGGCAACTACAAGACCCGCGCCGAGGTGAAG

TTCGAGGGCGACACCCTGGTGAACCGCATCGAGCTGAAGGGCATCGACTT

CAAGGAGGACGGCAACATCCTGGGGCACAAGCTGGAGTACAACTACAACA

GCCACAACGTCTATATCATGGCCGACAAGCAGAAGAACGGCATCAAGGTG

AACTTCAAGATCCGCCACAACATCGAGGACGGCAGCGTGCAGCTCGCCGA

CCACTACCAGCAGAACACCCCCATCGGCGACGGCCCCGTGCTGCTGCCCG

ACAACCACTACCTGAGCACCCAGTCCGCCCTGAGCAAAGACCCCAACGAG

AAGCGCGATCACATGGTCCTGCTGGAGTTCGTGACCGCCGCCGGGATCAC

TCTCGGCATGGACGAGCTGTACAAGTAAAGCGGCCGCAATTCACTCCTCA

GGTGCAGGCTGCCTATCAGAAGGTGGTGGCTGGTGTGGCCAATGCCCTTG

CTCACCATGGTGAAGGGTGGGCGCGCCGACCCAGCTTTCTTGTACAAAGT

GGTTGATCTAGAGGGCCCGCGGTTCGAAGGTAAGCCTATCCCTAACCCTC

TCCTCGGTCTCGATTCTACGCGTACCGGTCATCATCACCATCACCATTGA

GTTTATCCATCACACTGGCGGCCGCTCGAGGGGGATCCACTAGTTCTAGA

GCGGCCGCTTCGAGCAGACATGATAAGATACATTGATGAGTTTGGACAAA

CCACAACTAGAATGCAGTGAAAAAAATGCTTTATTTGTGAAATTTGTGAT

GCTATTGCTTTATTTGTAACCATTATAAGCTGCAATAAACAAGTTAACAA

CAACAATTGCATTCATTTTATGTTTCAGGTTCAGGGGGAGATGTGGGAGG

TTTTTTAAAGCAAGTAAAACCTCTACAAATGTGGTAAAATCGCGAAGACA

AGTGGCCAGAAAGGGCCGGGCGGAAGAGCCCTCGACAGACTGACCCTAAA

GCAAGACGAGGCAAGGCCAGTCCAGAATACCAGAGTGG

Nanog

Nanog gRNA:

(SEQ ID NO: 20)

GTATGAGACTTACGCAACATC

Nanog genomic region:

(SEQ ID NO: 21)

GCAGCCTTACGTACAGTTGCAGCAAAACTTCTCTGCCAGTGATTTGGAGG

TGAATTTGGAAGCCACTAGGGAAAGCCATGCGCATTTTAGCACCCCACAA

GCCTTGGAATTATTCCTGAACTACTCTGTGACTCCACCAGGTGAAATATG

AGACTTACGCAACATCTGGGCTTAAAGTCAGGGCAAAGCCAGGTTCCTTC

CTTCTTCCAAATATTTTCATAttttttttaaagatttatttattcattat

atgtaagtacactgtagctgtcttca

Nanog GFP fusion:

(SEQ ID NO: 22)

GCAGCCTTACGTACAGTTGCAGCAAAACTTCTCTGCCAGTGATTTGGAGG

TGAATTTGGAAGCCACTAGGGAAAGCCATGCGCATTTTAGCACCCCACAA

GCCTTGGAATTATTCCTGAACTACTCTGTGACTCCACCAGGTGAAATAGT

GAGCAAGGGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCTGGTCGAGC

TGGACGGCGACGTAAACGGCCACAAGTTCAGCGTGTCCGGCGAGGGCGAG

GGCGATGCCACCTACGGCAAGCTGACCCTGAAGTTCATCTGCACCACCGG

CAAGCTGCCCGTGCCCTGGCCCACCCTCGTGACCACCCTGACCTACGGCG

TGCAGTGCTTCAGCCGCTACCCCGACCACATGAAGCAGCACGACTTCTTC

AAGTCCGCCATGCCCGAAGGCTACGTCCAGGAGCGCACCATCTTCTTCAA

GGACGACGGCAACTACAAGACCCGCGCCGAGGTGAAGTTCGAGGGCGACA

CCCTGGTGAACCGCATCGAGCTGAAGGGCATCGACTTCAAGGAGGACGGC

AACATCCTGGGGCACAAGCTGGAGTACAACTACAACAGCCACAACGTCTA

TATCATGGCCGACAAGCAGAAGAACGGCATCAAGGTGAACTTCAAGATCC

GCCACAACATCGAGGACGGCAGCGTGCAGCTCGCCGACCACTACCAGCAG

AACACCCCCATCGGCGACGGCCCCGTGCTGCTGCCCGACAACCACTACCT

GAGCACCCAGTCCGCCCTGAGCAAAGACCCCAACGAGAAGCGCGATCACA

TGGTCCTGCTGGAGTTCGTGACCGCCGCCGGGATCACTCTCGGCATGGAC

GAGCTGTACAAGTAAAGCGGCCGCAATTCACTCCTCAAACATCTGGGCTT

AAAGTCAGGGCAAAGCCAGGTTCCTTCCTTCTTCCAAATATTTTCATAtt

ttttttaaagatttatttattcattatatgtaagtacactgtagctgtct

tca

Rpp25

Rpp25 gRNA:

(SEQ ID NO: 23)

GCTCAGAGGCGAGAATTCTC

Rpp25 genomic region:

(SEQ ID NO: 24)

CTCAGTCCTGGTCCTTCGTCCCCTCCTACGGTGTCGACGTCCAAGAGGAG

CCTGGGGGAATCTGCTGCTGAAGAAGGCACCGCTAAGCGGTCTCAGCCTG

AGCCAGAGGCTGAGAATGAGGACAGGACCGCCTGAGAATTCTCGCCTCTG

AGCCACCCAGACCGACTGAATCATATATCTTCAACACTCCTGCATACCTT

TCAACACACGCACCTTTCATACCTGGGTTTTAAGGGGCCCATGTTCCTG

Rpp25 GFP fusion:

(SEQ ID NO: 25)

CTCAGTCCTGGTCCTTCGTCCCCTCCTACGGTGTCGACGTCCAAGAGGAG

CCTGGGGGAATCTGCTGCTGAAGAAGGCACCGCTAAGCGGTCTCAGCCTG

AGCCAGAGGCTGAGAATGAGGACAGGACCGCCGTGAGCAAGGGCGAGGAG

CTGTTCACCGGGGTGGTGCCCATCCTGGTCGAGCTGGACGGCGACGTAAA

CGGCCACAAGTTCAGCGTGTCCGGCGAGGGCGAGGGCGATGCCACCTACG

GCAAGCTGACCCTGAAGTTCATCTGCACCACCGGCAAGCTGCCCGTGCCC

TGGCCCACCCTCGTGACCACCCTGACCTACGGCGTGCAGTGCTTCAGCCG

CTACCCCGACCACATGAAGCAGCACGACTTCTTCAAGTCCGCCATGCCCG

AAGGCTACGTCCAGGAGCGCACCATCTTCTTCAAGGACGACGGCAACTAC

AAGACCCGCGCCGAGGTGAAGTTCGAGGGCGACACCCTGGTGAACCGCAT

CGAGCTGAAGGGCATCGACTTCAAGGAGGACGGCAACATCCTGGGGCACA

AGCTGGAGTACAACTACAACAGCCACAACGTCTATATCATGGCCGACAAG

CAGAAGAACGGCATCAAGGTGAACTTCAAGATCCGCCACAACATCGAGGA

CGGCAGCGTGCAGCTCGCCGACCACTACCAGCAGAACACCCCCATCGGCG

ACGGCCCCGTGCTGCTGCCCGACAACCACTACCTGAGCACCCAGTCCGCC

CTGAGCAAAGACCCCAACGAGAAGCGCGATCACATGGTCCTGCTGGAGTT

CGTGACCGCCGCCGGGATCACTCTCGGCATGGACGAGCTGTACAAGTAAA

GCGGCCGCAATTCACTCCTCACTGAGCCACCCAGACCGACTGAATCATAT

ATCTTCAACACTCCTGCATACCTTTCAACACACGCACCTTTCATACCTGG

GTTTTAAGGGGCCCATGTTCCTG

MERA gRNA Library Design

We used the following algorithm to design gRNAs:

1. Determine region of interest for guide RNA design

2. Find all GG sequences on both the forward and reverse strand

3. Design guide RNA in the following format. Guide RNAs should have 19-20 bp of homology to the genome immediately preceding the NGG “PAM” sequence:

a. If the genome sequence is GNNNNNNNNNNNNNNNNNNN NGG (GN₁₉NGG), the guide RNA sequence should be GNNNNNNNNNNNNNNNNNNN (GN₁₉)

b. If a is not satisfied but GNNNNNNNNNNNNNNNNNN NGG (GN₁₈NGG) is satisfied, the guide RNA sequence should be GNNNNNNNNNNNNNNNNNN (GN₁₈)

c. If a and b are not satisfied, the guide RNA sequence should be GNNNNNNNNNNNNNNNNNNNN (GN₂₀) where the genomic sequence is NNNNNNNNNNNNNNNNNNNN NGG (N₂₀NGG)—it does not matter if the first G is in the genome.

4. For guide RNA library design, each guide RNA sequence should be placed in the following template, which will be 98-100 bp depending on guide RNA class a-c. We ordered gRNA libraries of 3918 members from LC Sciences. TTATATATCTTGTGGAAAGGACGAAACACC[GN_18-20]GTTTAAGAGCTATGCTGGAA ACAGCATAGCAAGTTTAAATAAGGCTAGT (SEQ ID NO: 26)

All libraries contained 10 gRNAs targeting the GFP open reading frame to serve as positive controls:

10 Positive Control gRNAs Targeting GFP:

1.

(SEQ ID NO: 27)

TTATATATCTTGTGGAAAGGACGAAACACCGGGCGAGGAGCTGTTCACCG

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

2.

(SEQ ID NO: 28)

TTATATATCTTGTGGAAAGGACGAAACACCGACCAGGATGGGCACCACCC

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

3.

(SEQ ID NO: 29)

TTATATATCTTGTGGAAAGGACGAAACACCGAGCTGGACGGCGACGTAAA

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

4.

(SEQ ID NO: 30)

TTATATATCTTGTGGAAAGGACGAAACACCGGCATCGCCCTCGCCCTCGC

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

5.

(SEQ ID NO: 31)

TTATATATCTTGTGGAAAGGACGAAACACCGCTTCAGGGTCAGCTTGCCG

TGTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

6.

(SEQ ID NO: 32)

TTATATATCTTGTGGAAAGGACGAAACACCGGGCACGGGCAGCTTGCCGG

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

7.

(SEQ ID NO: 33)

TTATATATCTTGTGGAAAGGACGAAACACCGGTCAGGGTGGTCACGAGGG

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

8.

(SEQ ID NO: 34)

TTATATATCTTGTGGAAAGGACGAAACACCGCTTCATGTGGTCGGGGTAG

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

9.

(SEQ ID NO: 35)

TTATATATCTTGTGGAAAGGACGAAACACCGACGTAGCCTTCGGGCATGG

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

10.

(SEQ ID NO: 36)

TTATATATCTTGTGGAAAGGACGAAACACCGGAGCGCACCATCTTCTTCA

GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGCTAGT

MERA Screening

To amplify the gRNA pool for electroporation into target cells, we use NEBNext 2× MasterMix, as it is optimized for maintaining diversity in pooled PCRs. We use a single PCR to add HDRs on each side of the gRNA library:

0.25 uL gRNA library (1% of the library) in a 500 uL NEBNext reaction (250 uL 2×NEBNext MasterMix, 1.25 uL of each primer or 25 of 10 uM primer mix)

gRNALib_HDR_fw

(SEQ ID NO: 37)

TGTTTTAAAATGGACTATCATATGCTTACCGTAACTTGAAAGTATTTCGA

TTTCTTGGCTTTATATATCTTGTGGAAAGGACGAAACACC

gRNALib_HDR_rv

(SEQ ID NO: 38)

CTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTAAACTTG

CTATGCTGTT

35 cycles of 3-step PCR (98 for 10s, 62 for 30s, 72 for 30s).

We test ˜2 uL, expecting 189 bp band.

We MinElute PCR purify the product using 2 columns (250 uL of PCR product per column).

The expected gRNA with HDR sequence is:

(SEQ ID NO: 39)

TGTTTTAAAATGGACTATCATATGCTTACCGTAACTTGAAAGTATTTCGA

TTTCTTGGCTTTATATATCTTGTGGAAAGGACGAAACACCGGGCGAGGAG

CTGTTCACCGGTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAAT

AAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAG

We then co-electroporate mESC that have ROSA dummy gRNA and locus-specific GFP with Cas9, guide RNA that cleaves the dummy gRNA, and purified homology arm-containing gRNA library. Library diversity is important here—we aim to have >10{circumflex over ( )}6 cells survive selection to have adequate diversity and thus we start with a ˜80% confluent 15 cm plate of mESC with:

25 ug p2T CBh Cas9 BlastR

25 ug p2T U6sgU6sgIhh-FE HygR (p2T sg60 HygR)

MinElute-purified gRNA library PCR product from 500 uL PCR reaction

Vacuum concentrate mix to <40 uL total volume and use 200 uL electroporation buffer We plated electroporated cells onto a 15-cm plate adding 7.5 uM Y-27632 to media for the 24 hours after electroporation to boost survival, which helps since library diversity is key. We transiently select with 10 ug/mL Blasticidin from 24-72 hours after electroporation (total of 48 hours of selection).

We expanded these cells, making sure to freeze down aliquots (usually passage beginning 4-6 days after electroporation) but maintain library diversity by not splitting at less than 1/10.

We then flow cytometrically isolated GFP(low) and GFP^neg cells using a FACSAria (BD Biosciences). We sorted between 1-2 weeks after electroporation. During the trypsinization that will be used for flow cytometry, we performed genomic DNA isolation on ˜⅓-⅕ of the cells to serve as the bulk genomic DNA control for gRNA incorporation. We performed a first sort where we collected one combined GFP(low)/neg population, grew these cells for 3-4 additional days in culture, then re-sorted, this time separating GFP(low) and GFP^neg cells when possible (for Tdgf1 and Zfp42 screens). After the second sort, we expanded the GFP(low) and GFP^neg pools for 3-4 additional days before collecting genomic DNA. If populations were still impure, they were sorted a third time, grown for 3-4 days, and then split for genomic DNA isolation.

We collected genomic DNA from bulk and sorted populations using Purelink Genomic DNA isolation kit (Life Technologies)

We then proceeded to a homemade, PCR-based library prep on all populations.

We use a 3-step PCR-based library prep.

PCR1 exclusively amplifies gRNAs in the ROSA locus in the population (bulk or sorted). This is important because even weeks after electroporation, unincorporated gRNA homology constructs can still be found in cells. We perform 15 cycles of PCR1 to enrich ROSA-locus gRNAs away from unincorporated ones. PCR2 and PCR3 add first half and then the entire Illumina paired-end sequencing primers. In PCR2, a sample barcode is introduced between the gRNA and the PE1 primer, which allows sample multiplexing in a MiSeq lane.

PCR1: We used up to 16 ug of genomic DNA in up to an 800 uL NEBNext reaction, keeping a ratio of <20 ng genomic DNA per uL of PCR reaction to avoid template “poisoning”. We typically used ½ of the genomic DNA isolated from cells We used the following primers at 500 nM:

082214_gRNA_upstream_fw

(SEQ ID NO: 40)

	TTGTGGAAAGGACGAAACACC
	8. 091514_ROSA_downstream_rv

(SEQ ID NO: 9)

GGGAGGGGAGTGTTGCAATA

We performed 15 cycles of 3-step PCR (98 for 10s, 60 for 30s, 72 for 30s).

The product is 576 bp (and a mix of gRNAs):

(SEQ ID NO: 41)

TTGTGGAAAGGACGAAACACC[GN18-20]

GTTTAAGAGCTATGCTGGAAACAGCATA

GCAAGTTTAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGA

GTCGGTGCTTTTTTGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCT

AGTCCGTTTTTAGCGCGTGCGCCAATTCTGCAGACAAATGGCTCTAGAGG

TACGGCCGCTTCGAGCAGACATGATAAGATACATTGATGAGTTTGGACAA

ACCACAACTAGAATGCAGTGAAAAAAATGCTTTATTTGTGAAATTTGTGA

TGCTATTGCTTTATTTGTAACCATTATAAGCTGCAATAAACAAGTTAACA

ACAACAATTGCATTCATTTTATGTTTCAGGTTCAGGGGGAGATGTGGGAG

GTTTTTTAAAGCAAGTAAAACCTCTACAAATGTGGTAAAATCGCGATGCA

GATCACGAGGGAAGAGGGGGAAGGGATTCTCCCAGGCCCAGGGCGGTCCT

CAGAAGCCAGGAGGCAGCAGAGAACTCCCAGAAAGGTATTGCAACACTCC

CCTCCC

We performed PCR purification in a single column per sample, eluting in 45 uL of dH2O. We used 0.1 uL of this purified product in a 20 uL qPCR with:

082214_gRNA_upstream_fw

(SEQ ID NO: 42)

	TTGTGGAAAGGACGAAACACC
	020515_gRNA_qPCR_rv

(SEQ ID NO: 43)

GCCTTATTTAAACTTGCTATGCTGT

To determine cycle count for PCR2, we divided the Ct by 2. Typical Ct values are 9-14.

PCR 2:

We used 23 uL of sample in a 50 uL NEBNext reaction. We used the following primers at 500 nM and different barcoded PE1 primer for each sample to be multiplexed:

101714_gRNAPE1_BcO

(SEQ ID NO: 44)

	CTCTTTCCCTACACGACGCTCTTCCGATCTaactc
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcP

(SEQ ID NO: 45)

	CTCTTTCCCTACACGACGCTCTTCCGATCTctgga
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcQ

(SEQ ID NO: 46)

	CTCTTTCCCTACACGACGCTCTTCCGATCTggact
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcR

(SEQ ID NO: 47)

	CTCTTTCCCTACACGACGCTCTTCCGATCTtctgc
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcS

(SEQ ID NO: 48)

	CTCTTTCCCTACACGACGCTCTTCCGATCTaaccg
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcT

(SEQ ID NO: 49)

	CTCTTTCCCTACACGACGCTCTTCCGATCTctctg
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcU

(SEQ ID NO: 50)

	CTCTTTCCCTACACGACGCTCTTCCGATCTggtaa
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcV

(SEQ ID NO: 51)

	CTCTTTCCCTACACGACGCTCTTCCGATCTaagct
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcW

(SEQ ID NO: 52)

	CTCTTTCCCTACACGACGCTCTTCCGATCTtcgtc
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcX

(SEQ ID NO: 53)

	CTCTTTCCCTACACGACGCTCTTCCGATCTccaat
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcY

(SEQ ID NO: 54)

	CTCTTTCCCTACACGACGCTCTTCCGATCTgcgta
	TTGTGGAAAGGACGAAACACC
	101714_gRNAPE1_BcZ

(SEQ ID NO: 55)

	CTCTTTCCCTACACGACGCTCTTCCGATCTtgagc
	TTGTGGAAAGGACGAAACACC
	9. 010715_LibrarygRNA_PE2

(SEQ ID NO: 56)

	CATTCCTGCTGAACCGCTCTTCCGATCTGCCTTATT
	TAAACTTGCTATGCTGT

We performed 3-step PCR (98 for 10s, 60 for 30s, 72 for 30s) using the number of cycles determined from the qPCR.

An example product with Barcode 0 is 150 bp:

(SEQ ID NO: 57)

CTCTTTCCCTACACGACGCTCTTCCGATCTaactcTTGTGGAAAGGACGN

18-20GTTTAAGAGCTATGCTGGAAACAGCATAGCAAGTTTAAATAAGGC

AGATCGGAAGAGCGGTTCAGCAGGAATG

We performed PCR purification in a single column per sample, eluting in 45 uL of dH2O. We used 0.1 uL of this purified product in a 20 uL qPCR with:

082214_gRNA_upstream_fw

(SEQ ID NO: 42)

TTGTGGAAAGGACGAAACACC

020515_gRNA_qPCR_rv

(SEQ ID NO: 43)

GCCTTATTTAAACTTGCTATGCTGT

We used an equivalent number of cycles of PCR3 as the Ct count.

PCR 3:

Use 23 uL of sample in a 50 uL NEBNext reaction. Use the following primers at 500 nM:

061813_PE1

(SEQ ID NO: 58)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCT

061813_PE2

(SEQ ID NO: 59)

CAAGCAGAAGACGGCATACGAGATCGGTCTCGGCATTCCTGCTGAACCGC

TCTTCCGATCT

We performed 3-step PCR (98 for 10s, 65 for 30s, 72 for 30s) using the number of cycles determined from the qPCR.

We performed PCR purification in a single column per sample, eluting in 30 uL of Elution buffer. The expected product is: PE BcO (211 bp)

(SEQ ID NO: 60)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCTaactcTTGTGGAAAGGACGAAACACCGN18-20

GTTTAAGAGCTATGCTGGAAACAG

CATAGCAAGTTTAAATAAGGCAGATCGGAAGAGCGGTTCAGCAGGAATGC

CGAGACCGATCTCGTATGCCGTCTTCTGCTTG

Validation of MERA Hit gRNAs

We validated hit gRNAs from the MERA screens using the Self-Cloning CRISPRCas9 (scCRISPR) protocol we have recently developed (Arbab et al. Stem Cell Reports, 2015). The scCRISPR protocol enables testing of gRNAs simply by ordering a single oligonucleotide and PCR amplifying it to form a homology fragment. As a result, testing of gRNAs can be performed at <⅙ of the cost of constructing a conventional plasmid gRNA with only two hours of preparation from when oligonucleotides arrive to when they can be tested in mESCs.

Briefly, mESCs were co-transfected using Lipofectamine 3000 with a Cas9 plasmid, a self-cleaving gRNA plasmid, and a homology fragment to replace the cleaved gRNA plasmid with an gRNA sequence of interest. After transient drug selection and growth for 6-9 days after transfection, mESCs were analyzed for fluorescence using a BD Accuri C6 flow cytometer. The oligonucleotides to test Tdgf1 hit gRNAs, control gRNAs, as well as the scCRISPR stock primers are listed below:

gRNA_60bp_fw

(SEQ ID NO: 61)

TAACTTGAAAGTATTTCGATTTCTTGGCTTTATATATCTTGTGGAAAGGA

CGAAACACCG

gRNA_60bp_rv

(SEQ ID NO: 62)

GTTGATAACGGACTAGCCTTATTTAAACTTGCTATGCTGTTTCCAGCATA

GCTCTTAAAC

gRNAHDR_Ext_fw

(SEQ ID NO: 63)

ATTTCTTGGGTAGTTTGCAGTTTTAAAATTATGTTTTAAAATGGACTATC

ATATGCTTACCGTAACTTGAAAGTATTTCGATTTCTTGGC

gRNAHDR_Ext_rv

(SEQ ID NO: 64)

ATTTTAACTTGCTATTTCTAGCTCTAAAACAAAAAAGCACCGACTCGGTG

CCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTAAAC

gRNALib_HDRstep3_fw

(SEQ ID NO: 65)

CGATACAAGGCTGTTAGAGAGATAATTAGAATTAATTTGACTGTAAACAC

AAAGATATTAGTACAAAATACGTGACGTAGAAAGTAATAA

gRNAHDR_step3_rv

(SEQ ID NO: 66)

TCAATGTATCTTATCATGTCTGCTCGATTTTAACTTGCTATTTCTAGCTC

TAAAACAAAA

sgGFP4_60bp (positive control in GFP ORF)

(SEQ ID NO: 67)

TGGAAAGGACGAAACACCGGCATCGCCCTCGCCCTCGC

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU1_60bp

(SEQ ID NO: 68)

TGGAAAGGACGAAACACCGAGTGAGATCCAGGTGATCC

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU2_60bp

(SEQ ID NO: 69)

GGAAAGGACGAAACACCGCAAACAGCTCCATACCCAAG

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU3_60bp

(SEQ ID NO: 70)

TGGAAAGGACGAAACACCGCAAGCACCACCCTACTTGG

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU4_60bp

(SEQ ID NO: 71)

GGAAAGGACGAAACACCGCCATCCCCTGCCGGTCTACA

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU5_60bp

(SEQ ID NO: 72)

TGGAAAGGACGAAACACCGGGGGTGAGGGGCAATTGG

GTTTAAGAGCTATGCTGGAAACA

sgTdgfFU6_60bp

(SEQ ID NO: 73)

GGAAAGGACGAAACACCGAATTCCTGCTTCTTACCACG

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU8_60bp

(SEQ ID NO: 74)

TGGAAAGGACGAAACACC

GAAACAAAAGCCACAAACA GTTTAAGAGCTATGCTGGAAACA

sgTdgfFU9_60bp

(SEQ ID NO: 75)

GGAAAGGACGAAACACC

GATGGTTAGCATTCCGGGTGC GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU10_60bp

(SEQ ID NO: 76)

GGAAAGGACGAAACACC

GTTATGGGCAAGCTGTGTCCC GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU11_60bp

(SEQ ID NO: 77)

TGGAAAGGACGAAACACC GGGGAAGTGACTAAGACTG

GTTTAAGAGCTATGCTGGAAACA

sgTdgfFU12_60bp

(SEQ ID NO: 78)

TGGAAAGGACGAAACACC GCCATAGTTTTCCAGAAAG

GTTTAAGAGCTATGCTGGAAACA

sgTdgfFU13_60bp

(SEQ ID NO: 79)

GGAAAGGACGAAACACC

GTCTGAAGGAGACAATGACAA GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU14_60bp

(SEQ ID NO: 80)

GGAAAGGACGAAACACC GCCCATCCCCTGTCACTCT

GAGTTCGTTTAAGAGCTATGCTGGAAAC

sgTdgfFU15_60bp

(SEQ ID NO: 81)

GGAAAGGACGAAACACC

GATAATTCCTTTAGTCTACAT GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU16_60bp

(SEQ ID NO: 82)

TGGAAAGGACGAAACACC

GGGATTCTGGGAAAACATTG GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU17_60bp

(SEQ ID NO: 83)

GGAAAGGACGAAACACC

GCTCTCAGGACTTGTCCACAC GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU18_60bp

(SEQ ID NO: 84)

GGAAAGGACGAAACACC

GTTTCCTAAGCCCTAGCTGGA GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU19_60bp

(SEQ ID NO: 85)

GGAAAGGACGAAACACC

GTTAGAAATTAGTGCAGTGTT GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU20_60bp

(SEQ ID NO: 86)

GGAAAGGACGAAACACC GAATCTTCCCAATTGTCTCCT

GTTTAAGAGCTATGCTGGAAAC

sgTdgfFU21_60bp

(SEQ ID NO: 87)

TGGAAAGGACGAAACACC

GTCAACTAAACCAACACTTA GTTTAAGAGCTATGCTGGAAAC

TdgfFU25_60bp

(SEQ ID NO: 88)

GGAAAGGACGAAACACC

GTCTTTCAAAACAAGACCCAA GTTTAAGAGCTATGCTGGAAAC

TdgfFU26_60bp

(SEQ ID NO: 89)

GGAAAGGACGAAACACC

GTTTCTGTCCAGGACACTGAT GTTTAAGAGCTATGCTGGAAAC

TdgfFU27_60bp

(SEQ ID NO: 90)

GGAAAGGACGAAACACC

GAAAAAACTAGTCAACATTCC GTTTAAGAGCTATGCTGGAAAC

TdgfFU28_60bp

(SEQ ID NO: 91)

GGAAAGGACGAAACACC

GCTGTGAGCCCTTTGTAAGGA GTTTAAGAGCTATGCTGGAAAC

TdgfFU29_60bp

(SEQ ID NO: 92)

GGAAAGGACGAAACACC

GACTGCATCCACAGACTGGGC GTTTAAGAGCTATGCTGGAAAC

TdgfFU30_60bp

(SEQ ID NO: 93)

GGAAAGGACGAAACACC

GCCCTCACCCCCTAGGGTTTC GTTTAAGAGCTATGCTGGAAAC

sgTdgfCtlFU1_60bp

(SEQ ID NO: 94)

TGGAAAGGACGAAACACC GTTTCTTCATTGTCAGAGA

GTTTAAGAGCTATGCTGGAAACA

sgTdgfCtlFU2_60bp

(SEQ ID NO: 95)

GGAAAGGACGAAACACC GTTCTTCCTTGCTTTCTTTTG

GTTTAAGAGCTATGCTGGAAAC

TdgfFU43_60bp

(SEQ ID NO: 96)

TGGAAAGGACGAAACACC GAATTTTCCTTCATCATTTT

GTTTAAGAGCTATGCTGGAAAC

TdgfFU44_60bp

(SEQ ID NO: 97)

GGAAAGGACGAAACACC

GTTCAGCACAGCAGATGCTCT GTTTAAGAGCTATGCTGGAAAC

TdgfFU45_60bp

(SEQ ID NO: 98)

TGGAAAGGACGAAACACC

GAACAGCTTTTTAGATGTGA GTTTAAGAGCTATGCTGGAAAC

TdgfFU47_60bp

(SEQ ID NO: 99)

TGGAAAGGACGAAACACC

GATTTAGAATCCCCTAAGGA GTTTAAGAGCTATGCTGGAAAC

TdgfFU49_60bp

(SEQ ID NO: 100)

GGAAAGGACGAAACACC

GTCCGAGACTGCGGCTCTTAC GTTTAAGAGCTATGCTGGAAAC

TdgfFU50_60bp

(SEQ ID NO: 101)

GGAAAGGACGAAACACC

GTCGGATGACCTGAGTAGAGC GTTTAAGAGCTATGCTGGAAAC

Deep Sequencing of MERA Hotspots

We used the scCRISPR method to induce mutations at eight individual gRNA sites in separate batches of mESCs. We then pooled batches with gRNAs in each of four hotspots. GFP^pos and GFP^medium/- populations were flow cytometrically sorted from these five pools, all giving >90% purity. Genomic DNA was isolated from these populations using the Purelink Genomic DNA mini kit. PCR-based library prep was performed using four successive PCRs to amplify and add paired-end Illumina sequencing primers and multiplexing barcodes to desired genomic regions. Paired-end MiSeq was performed on the prepared libraries using 150+150 nt paired-end reads. Sequenced regions with Illumina paired-end primers and n's to denote multiplexing barcodes along with all primers used for this protocol are shown below:

Tdgf1 enhancer region

(SEQ ID NO: 102)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCTnnnnn GCCGAAAGGATGGGAGTACTAAGC

10.

(SEQ ID NO: 103)

ACGCCCAGTCTCTCTACTCCCCCCCACCCCCATCCCCTGTCACTCTGAGT

TTCCAGCCACTTTTCCAGTTCCTGAAACCCTAGGGGGTGAGGGGCAATTG

GTGGTGGTGGTGGGGGGGGGGAATCTGCTATTTCCGAGAAGGCTGGGCCT

CCTTCATTAACAAGCTAATGGCTGATTTCACTGAGACCTTGACA

TGGATGCAGGTCGAAAGGCCTnnnnnAGATCGGAAGAGCGGTTCAGCAGG

AATGCCGAGACCGATCTCGTATGCCGTCTTCTGCTTG

sgTdgfFU5_60bp

(SEQ ID NO: 104)

TGGAAAGGACGAAACACCGGGGGTGAGGGGCAATTGGGTTTAAGAGCTAT

GCTGGAAACA

11. TdgfFU30_60bp

(SEQ ID NO: 105)

GGAAAGGACGAAACACC

GCCCTCACCCCCTAGGGTTTC GTTTAAGAGCTATGCTGGAAAC

12. 042115_TdgfFU30_up_fw

(SEQ ID NO: 106)

ACGCCCAGTCTCTCTACTCCCC

13. 060915_TdgfFU30_dwn_rv

(SEQ ID NO: 107)

TGTCAAGGTCTCAGTGAAATCAGCCA

14. 042915_TdgfFU30_ScaRO_fw

(SEQ ID NO: 108)

CGAAAGGATGGGAGTACTAAGCT

ACGCCCAGTCTCTCTACTCCCC

15. 060915_TdgfFU30_IntPri_rv

(SEQ ID NO: 109)

GGCCTTTCGACCTGCATCCA

TGTCAAGGTCTCAGTGAAATCAGCCA

16. Tdgf1 URE region

(SEQ ID NO: 110)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCTnnnnn GCCGAAAGGATGGGAGTACTAAGCT

17.

(SEQ ID NO: 111)

CCAAATGGGCCAAGCAAAACACAGCCAGTGAGACAGGCCTGCACCCGGAA

TGCTAACCATCAAGAGTGGATTCCGGGAGAGGGGCAACCTGGTTCAACCA

GCGACTCACAGATGAGACTGTGAGCCCTTTGTAAGGAAGGACTGTCTGTA

GACTGAATGCGAGCTGAGCTCACGTGTCACTGGCCACAGAGGTCCAGCCC

AGTCTGTGGATGCAGTGGAGTCCAGGAAGGGCTTTCTCTGGCTTTTGTAG

TGGGACCG

TGGATGCAGGTCGAAAGGCCTnnnnnAGATCGGAAGAGCGGTTCAGCAGG

AATGCCGAGACCGATCTCGTATGCCGTCTTCTGCTTG

sgTdgfFU9_60bp

(SEQ ID NO: 112)

GGAAAGGACGAAACACC

GATGGTTAGCATTCCGGGTGC GTTTAAGAGCTATGCTGGAAAC

TdgfFU29_60bp

(SEQ ID NO: 113)

GGAAAGGACGAAACACC

GACTGCATCCACAGACTGGGC GTTTAAGAGCTATGCTGGAAAC

060915_TdgfFU9_up_fw

(SEQ ID NO: 114)

CCAAATGGGCCAAGCAAAACACAG

021915_TdgfFU9_dwn_rv

(SEQ ID NO: 115)

CGGTCCCACTACAAAAGCCAGAGAA

060915_TdgfFU9_ScaRO_fw

(SEQ ID NO: 116)

CGAAAGGATGGGAGTACTAAGCT

CCAAATGGGCCAAGCAAAACACAG

030715_TdgfFU9_IntPri_rv

(SEQ ID NO: 117)

GGCCTTTCGACCTGCATCCACGGTCCCACTACAAAAGCCAGAGAA

Zfp42 Enhancer region

(SEQ ID NO: 118)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCTnnnnn GCCGAAAGGATGGGAGTACTAAGC

AGCGGGGTGGTAGCTCACAAGACCAGGTAGGACGGATATGGCTTTGCGCA

GGCAATTAGTTTCTAAGAGCTCTCATTCAGTGAATGTGAACAGTGCCTTT

TACAAAAGATGCTTGGGTGAAAGGACAGAGGAAGGTCGAGGGGGTGTGGC

TTTGAGCAGGCAATTGGCTCCCCAGAGCTCAGATTTTATGAATGTGAACA

GTACCCTTTTAAAAAGAGGCTAGGGTGTAGGCA

TGGATGCAGGTCGAAAGGCCTnnnnnAGATCGGAAGAGCGGTTCAGCAGG

AATGCCGAGACCGATCTCGTATGCCGTCTTCTGCTTG

18. sgZfpFU11_60bp

(SEQ ID NO: 119)

TGGAAAGGACGAAACACC GTGGCTTTGAGCAGGCAAT

GTTTAAGAGCTATGCTGGAAACA

19. sgZfpFU36_60bp

(SEQ ID NO: 120)

TGGAAAGGACGAAACACC

GGACGGATATGGCTTTGCGC GTTTAAGAGCTATGCTGGAAAC

20. 060915_ZfpFU11_dwn_rv

(SEQ ID NO: 121)

TGCCTACACCCTAGCCTCTTTT

21. 060915_ZfpFU11_IntPri_rv

(SEQ ID NO: 122)

GGCCTTTCGACCTGCATCCA

TGCCTACACCCTAGCCTCTTTT

22. 061815_ZfpFU11_up_fw

(SEQ ID NO: 175)

TGGTAGCTCACAAGACCAGGT

23. 061815_ZfpFU11_ScaRO_fw

(SEQ ID NO: 123)

CGAAAGGATGGGAGTACTAAGCT

TGGTAGCTCACAAGACCAGGT

Zfp42 URE region

(SEQ ID NO: 124)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCTnnnnn GCCGAAAGGATGGGAGTACTAAGC

tggtgctcagacatacatgaaggcaaaatgtcatatacttaaaaaaaaaa

AAtgaagacatagctcagtggctactcctccagaggaccctggtacccac

atggcagctctaaacaccagttccaggggatccaataccctcACATGCAG

ACAAAACCAATGCACATAAAATAAAAACTAAAACACTAGAAAGTATTCCA

AGTGTGACCCCTCAATACCTAGCCTCTTTTCCATGTCCTCTACCTTTGCT

ATTCCACCTACATCTTCGGGGAGAAGAGACAGAAGGGCCACGCTAGACAC

ATAAAATCCCATTTTCTAGGCCTCAAAATCCAGATAGGAACATCTTGTAA

CTTCTAGAGACTTTTTCTCGTGAAAGGAGCTCAGAGCAGACCCACCTTTA

CAAGGAGAGCCAACCTTACCCTTCTGAGTGCTCTGAGGCTCCAGTCTGAA

GAGCCTGCTCCCTAATCTCTGCATCTTCTGGTTGGATGCAGGTCGAAAGG

CCTnnnnnAGATCGGAAGAGCGGTTCAGCAGGAATGCCGAGACCGATCTC

GTATGCCGTCTTCTGCTTG

24. sgZfpFU4_60bp

(SEQ ID NO: 125)

TGGAAAGGACGAAACACC GCACTCAGAAGGGTAAGGT

GTTTAAGAGCTATGCTGGAAACA

25. sgZfpFU34_60bp

(SEQ ID NO: 126)

TGGAAAGGACGAAACACC GGGTACCAGGGTCCTCTGG

GTTTAAGAGCTATGCTGGAAACA

26. 042115_ZfpFU4_up_fw

(SEQ ID NO: 127)

tggtgctcagacatacatgaaggca

27. 042115_ZfpFU4_dwn_rv

(SEQ ID NO: 128)

ACCAGAAGATGCAGAGATTAGGGAGC

28. 042915_ZfpFU4_ScaRO_fw

(SEQ ID NO: 129)

CGAAAGGATGGGAGTACTAAGCTggtgctcagacatacatgaaggca

29. 042915_ZfpFU4_IntPri_rv

(SEQ ID NO: _)

GGCCTTTCGACCTGCATCCACCAGAAGATGCAGAGATTAGGGAGC

032014_PhrPE1_BcO

(SEQ ID NO: 130)

CTCTTTCCCTACACGACGCTCTTCCGATCTaactc

GCCGAAAGGATGGGAGTACTAAGCT

032014_PhrPE1_BcP

(SEQ ID NO: 131)

CTCTTTCCCTACACGACGCTCTTCCGATCTctggaGCCGAAAGGATGGGA

GTACTAAGCT

032014_PhrPE1_BcQ

(SEQ ID NO: 132)

CTCTTTCCCTACACGACGCTCTTCCGATCTggactGCCGAAAGGATGGGA

GTACTAAGCT

032014_PhrPE1_BcR

(SEQ ID NO: 133)

CTCTTTCCCTACACGACGCTCTTCCGATCTtctgc

GCCGAAAGGATGGGAGTACTAAGCT

061813_PhrPE1_BcS

(SEQ ID NO: 134)

CTCTTTCCCTACACGACGCTCTTCCGATCTaaccgGCCGAAAGGATGGGA

GTACTAAGCT

061813_PhrPE1_BcT

(SEQ ID NO: 135)

CTCTTTCCCTACACGACGCTCTTCCGATCTctctg

GCCGAAAGGATGGGAGTACTAAGCT

061813_PhrPE1_BcU

(SEQ ID NO: 136)

CTCTTTCCCTACACGACGCTCTTCCGATCTggtaaGCCGAAAGGATGGGA

GTACTAAGCT

061813_PhrPE1_BcV

(SEQ ID NO: 137)

CTCTTTCCCTACACGACGCTCTTCCGATCTaagct

GCCGAAAGGATGGGAGTACTAAGCT

021314_PhrPE1_BcW

(SEQ ID NO: 138)

CTCTTTCCCTACACGACGCTCTTCCGATCTtcgtc

GCCGAAAGGATGGGAGTACTAAGCT

021314_PhrPE1_BcX

(SEQ ID NO: 139)

CTCTTTCCCTACACGACGCTCTTCCGATCTccaat

GCCGAAAGGATGGGAGTACTAAGCT

021314_PhrPE1_BcY

(SEQ ID NO: 140)

CTCTTTCCCTACACGACGCTCTTCCGATCTgcgtaGCCGAAAGGATGGGA

GTACTAAGCT

021314_PhrPE1_BcZ

(SEQ ID NO: 141)

CTCTTTCCCTACACGACGCTCTTCCGATCTtgagcGCCGAAAGGATGGGA

GTACTAAGCT

040914_IntPriPE2_BcA

(SEQ ID NO: 142)

CATTCCTGCTGAACCGCTCTTCCGATCT ACATCAGGCCTTTCGACCTGC

ATCCA

040914_IntPriPE2_BcB

(SEQ ID NO: 143)

CATTCCTGCTGAACCGCTCTTCCGATCT GCCTAAGGCCTTTCGACCTGC

ATCCA

040914_IntPriPE2_BcC

(SEQ ID NO: 144)

CATTCCTGCTGAACCGCTCTTCCGATCT TGGTCAGGCCTTTCGACCTGC

ATCCA

040914_IntPriPE2_BcD

(SEQ ID NO: 145)

CATTCCTGCTGAACCGCTCTTCCGATCT CACTGAGGCCTTTCGACCTGC

ATCCA

PE1

(SEQ ID NO: 58)

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT

TCCGATCT

PE2

(SEQ ID NO: 59)

CAAGCAGAAGACGGCATACGAGATCGGTCTCGGCATTCCTGCTGAACCGC

TCTTCCG ATCT

Deletion and Repair of Tdgf1 Enhancer and URE Regions

Deletion and repair experiments were performed using scCRISPR. Clonal mESC lines with noted deletions and repair to wildtype were picked and sequence verified by Sanger sequencing. Repair was performed by cutting with a gRNA that will only cut in the deletion genotype and by co-electroporating a PCR-amplified repair template including the region to be replaced with 100-300 bp of homology on either end. Deletion and repair were performed both in Tdgf1-GFP mESCs as well as wildtype mESCs. In wildtype mESCs, RT-qPCR was performed using NEB M-MuLV Reverse Transcriptase and qPCR comparison with Actb and Gapdh as cDNA controls.

Tdgf1 enhancer deletion and repair:

chr9: 110,947,795-110,948,285 (wildtype genotype)

30.

(SEQ ID NO: 146)

TCCTGGTAAATAACTGAGCCCTGAAATTGCTTGCTTACTCTGTGAATGCT

TCCACAATGTTTTCCCAGAATCCCACCCCTGAACCTTCGTGTGTACTGAG

CCCCTCTCTCAGGACTTGTCCACACTGGCTGCAGCAAACATCTGTAGCTT

GGCGTCTGCTGACTCCTGATGTGAGAAATTCGTATTTGGTTGTTTGGATT

TGCATGTTTCGCTGTGTTACTGGTATCTGTATTCAGTTTCCTAAGCCCTA

GCTGGAGGGTCTGACTGTGCCCTAGAACTGGCTTCTTTGGTCCTTTGCTC

TGGAAAGTCCTTCCGGGCACCTTTCCTCCCAAAGCTGGCTTCCAGCTGTG

GCCTTGAACTGGAAAGCGCAGAGAGATAACGACTGCTCCACCCCAGAGTT

GATCTTTACTAACACTGGATTGTAAACTGCTTATTTTTTTGTCTTTCACC

CCACTCCCCCTGGTTCCCAATGGGAGGTCTAAGCCTGCTCT

Deletion genotype:

31.

(SEQ ID NO: 147)

TCCTGGTAAATAACTGAGCCCTGAAATTGCTTGCTTACTCTGTGAATGCT

TCCACAACACTGGCTGCAGCAAACATCTGTAGCTTGGCGTCTGCTGACTC

CTGATGTGAGAAATTCGTATTTGGTTGTTTGGATTTGCATGTTTCGCTGT

GTTACTGGTATCTGTATTCAGTTTCCTAAGCCCTAGCTGGAGGGTCTGAC

TGTGCCCTAGAACTGGCTTCTTTGGTCCTTTGCTCTGGAAAGTCCTTCCG

GGCACCTTTCCTCCCAAAGCTGGCTTCCAGCTGTGGCCTTGAACTGGAAA

GCGCAGAGAGATAACGACTGCTCCACCCCAGAGTTGATCTTTACTAACAC

TGGATTGTAAACTGCTTATTTTTTTGTCTTTCACCCCACTCCCCCTGGTT

CCCAATGGGAGGTCTAAGCCTGCTCT

gRNAs used for deletion:

sgTdgfFU16_60bp

(SEQ ID NO: 82)

	TGGAAAGGACGAAACACC
	GGGATTCTGGGAAAACATTG GTTTAAGAGCTATGCTGGAAAC
	sgTdgfFU17_60bp

(SEQ ID NO: 83)

	GGAAAGGACGAAACACC
	GCTCTCAGGACTTGTCCACAC GTTTAAGAGCTATGCTGGAAAC

gRNA used for repair to wildtype:

(SEQ ID NO: 148)

GGAAAGGACGAAACACCGTTTGCTGCAGCCAGTGTTG GTTTAAGAGCT

ATGCTGGAAACA

Primers used for homology-directed repair template and sequencing:

TdgfFU16_up_fw

(SEQ ID NO: 149)

	TCCTGGTAAATAACTGAGCCCTGAAAT
	TdgfFU16_dwn_rv

(SEQ ID NO: 150)

AGAGCAGGCTTAGACCTCCCA

• • 32. Tdgf1 URE deletion and repair:

33. chr9: 110,933,942-110,934,053 (wildtype region)

34.

(SEQ ID NO: 151)

CCCCTTGGATGTGAGAACCTCAGGGTTCCAAGGACTCTCTTCTGGGAGTC

TGCCCAACTGCAAAAGGCTAGGTGGCTGACACTTGGAGATGGGGGTGGGA

GCAGATGCCACAGTCTTTTGACATGCCCACCAAAAGGCCATTTGGAAATA

AAGCTGCTTTGGTTGCCAGCAGAGCTCTTGTCTCAGAGGGGACCCTGGCA

GATGGCGGCGCGCCTGTTATCACGGGCATATCCCTGCTGATGTTCTTCCT

TCTTCGAAATAGAGCGTTTATTCAGCTCCAATTTGTTACCATGGGTTGTC

CCAAAATGATGAAGGAAAATTCAAGAGACTGCCAGGGGCCAGTTGGATTT

GAAACATTTGTATTCAGCACAGCAGATGCTCTCGGCTACAGAGAACAGCT

TTT

• • 35.
  • 36. Deletion genotype:

(SEQ ID NO: 152)

CCCCTTGGATGTGAGAACCTCAGGGTTCCAAGGACTCTCTTCTGGGAGTC

TGCCCAACTGCAAAAGGCTAGGTGGCTGACACTTGGAGATGGGGGTGGGA

GCAGATGCCACAGTCTTTTGACATGCCCACCAAAAGGCCATTTGGAAATA

AAGCTGCTTTGGTTGCCAGCAGAGCTCTTGTCTCAGAGGGGACCCTGGCA

GATGGCGGCGCGCCTGTTATTGAAGGAAAATTCAAGAGACTGCCAGGGGC

CAGTTGGATTTGAAACATTTGTATTCAGCACAGCAGATGCTCTCGGCTAC

AGAGAACAGCTTTT

gRNAs used for deletion:

TdgfFU39_60bp

(SEQ ID NO: 153)

	TGGAAAGGACGAAACACCGGCGGCGCGCCTGTTATCAC
	GTTTAAGAGCTATGCTGGAAAC
	TdgfFU40_60bp

(SEQ ID NO: 154)

	TGGAAAGGACGAAACACC
	GGGTTGTCCCAAAATGATGA GTTTAAGAGCTATGCTGGAAAC

gRNA used for repair to wildtype:

(SEQ ID NO: 155)

	GGAAAGGACGAAACACC GTTGAATTTTCCTTCAATAAC
	GTTTAAGAGCTATGCTGGAAAC

Primers used for homology-directed repair and sequencing:

TdgfFU37_up_fw

(SEQ ID NO: 156)

	CCCCTTGGATGTGAGAACCTC
	TdgfFU37_dwn_rv

(SEQ ID NO: 157)

AAAAGCTGTTCTCTGTAGCCGAGAG

RT-qPCR primers:

Gapdh_qPCRfw

(SEQ ID NO: 158)

	TTGATGGCAACAATCTCCAC
	Gapdh_qPCRrv

(SEQ ID NO: 159)

	CGTCCCGTAGACAAAATGGT
	Actb_qPCRfw

(SEQ ID NO: 160)

	ATGGAGGGGAATACAGCCC
	Actb_qPCRrv

(SEQ ID NO: 161)

	TTCTTTGCAGCTCCTTCGTT
	Tdgf1_qPCR1_fw

(SEQ ID NO: 162)

	CCTCCAACGTTTTTACGAGC
	Tdgf1_qPCR1_rv

(SEQ ID NO: 163)

	GGTCCAAATTCAAACGCACT
	Tdgf1_qPCR2_fw

(SEQ ID NO: 164)

	TTTTACGAGCCGTCGAAGAT
	Tdgf1_qPCR2_rv

(SEQ ID NO: 165)

TCTGATGGCAAGGTCTCTCC

Analysis of Mutation and Deletion Efficiency in the Tdgf1-Adjacent Lrrc2 Promoter Region

scCRISPR was used to perform mutation and deletion as noted in the text.

Lrrc2 promoter region:

(SEQ ID NO: 166)

atgtgggtcccagtagtcatattccacctggtgcctctggcctccgagtc

atctgactggcccTTAGTGAACTTTgtggttagtcatgttgtgtgcacag

tcaggaactagagagaaggctagtattcctagtgacttagtgtccgggat

gttgccactgccactcacagtcagaatagggcttttctactcagttaaac

actctcacagataccccagcagagccttccttaggggattctaaatccag

tcaaattcacagtgaagatgaaccctcCAGTGCTCCGTAGCTGGGGCTGC

TGCGTAGCACATTTGCACACCGTGAGGACCATCAGTAGCCCAGGTTGGCC

GGAGTTTTAGCCTGCAGATGTGAGGCCACAAGCCAGGAGCCCTTGCCGCT

TGCTCTCCCACCAGCGAGAACCATTTAGTAATGTCCATGGAGATGGAAGG

CGGTGTTTATAACATTCATTCTTCTCTACATACATCGTGAAGGGAGGGAG

GTCAGGTGAGGGACATCAGGGATTTTCCTGGAAGAAAGCTTTGCTGGGCG

TTCATGTGACTCATGTTCCAGTAAGAGCCGCAGTCTCGGATGACCTGAGT

AGAGCAGGGTTATCTGGATGTGCTTGTGGGTGGAACCCCTTTGGAAGGGA

GGTAGGTAAGGGTGGGTGTATCCTCGCAGAGACGTAGAACTTCTGTGTGT

GGATGATGTTTCATCAGAGACTGTGAGCCGAGCCGGATGCTAAACAcagt

agttctcaaccttcctaatgctacaaccctcatgctgtggtgacccccag

ccataca

gRNAs:

TdgfFU46_60bp

(SEQ ID NO: 167)

	GGAAAGGACGAAACACC
	GCTCTGCTGGGGTATCTGTGA GTTTAAGAGCTATGCTGGAAAC

Negative in MERA whose PAM sequence is actually NAG in the Tdgf1-GFP mESC line.

TdgfFU47_60bp

(SEQ ID NO: 168)

	TGGAAAGGACGAAACACC
	GATTTAGAATCCCCTAAGGA GTTTAAGAGCTATGCTGGAAAC

Positive in MERA

TdgfFU49_60bp

(SEQ ID NO: 169)

	GGAAAGGACGAAACACC
	GTCCGAGACTGCGGCTCTTAC GTTTAAGAGCTATGCTGGAAAC

Positive in MERA

TdgfFU50_60bp

(SEQ ID NO: 170)

	GGAAAGGACGAAACACC
	GTCGGATGACCTGAGTAGAGC GTTTAAGAGCTATGCTGGAAAC

Negative in MERA

PCR primers:

TdgfFU46_up_fw

(SEQ ID NO: 171)

	cctctggcctccgagtcatctga
	TdgfFU50_dwn_rv

(SEQ ID NO: 172)

TCATCCACACACAGAAGTTCTACGT

Results and Conclusions

To enable the efficient targeting of precisely one regulatory element per cell, we devised a strategy that ensures that only one gRNA can be expressed per cell and that allows gRNA libraries to be used without any molecular cloning into a delivery vector. We integrated a single copy of integrated a single copy of the gRNA expression construct (a U6 promoter driving expression of a dummy gRNA hairpin) into the universally accessible ROSA locus of mESCs using CRISPR/Cas9-mediated homologous recombination (FIG. 1A). We then use CRISPR/Cas9-mediated homologous recombination to replace the dummy gRNA with a gRNA from our library. We use PCR to add 79-90 bp homology arms to our gRNA library, as we found that longer homology arms increase background cutting of unintegrated gRNA PCR fragments (data not shown). We then introduce the pool of gRNA homology fragments into cells along with Cas9 and a gRNA plasmid that induces a DSB at the dummy gRNA site. In a substantial fraction of cells (˜30%), the dummy gRNA is repaired by homologous recombination, creating a functional gRNA expression construct targeting a single genomic site from the library (data not shown). It is random chance which gRNA is integrated in each cell, allowing a pooled screen in which each cell expresses only one gRNA. Of note, the genomic integration-based gRNA screening platform used in MERA could also be applied to other CRISPR-based high-throughput screens as long as the cell line undergoes homologous recombination at appreciable frequency, and it could be modified to achieve expression of any set number of gRNAs per cell for combinatorial screening. While the integration-based approach is thus ill-suited to in vivo screens or screens in cells with limited homologous recombination, it provides an alternative to lentiviral screening that substantially reduces the time, effort, and cost involved in CRISPR library screening for applicable cell lines such as ESCs.

We have recently reported a method enabling efficient CRISPR-mediated knock-in transgene insertion that uses PCR-amplified homology arms, removing the time-consuming step of cloning homology arm vectors (Arbab et al. Stem Cell Reports, 2015). We used this procedure to construct GFP knock-in alleles at four genes with mESC-specific expression, Nanog, Rpp25, Tdgf1, and Zfp42. We chose to focus on mESC-specific genes because the networks of mESC gene regulation and epigenetic state of mESC regulatory elements are among the best understood of any cell type^{29, 30}. Two of the genes (Nanog and Rpp25) were constructed as C-terminal GFP fusion proteins, and two (Tdgf1 and Zfp42) had their open reading frames replaced with GFP to compare these two approaches. All knock-in lines, upon clonal selection, had robust GFP expression, although GFP levels were highest in Tdgf1^GFP and Zfp42^GFP lines (FIG. 1B, data not shown). The Nanog^GFP line always contained ˜20% GFP′ cells, presumably due to known heterogeneous Nanog expression in mESCs (data not shown)³¹.

We synthesized four gRNA libraries, each with 3908 gRNAs tiling cis-regulatory regions of Nanog, Rpp25, Tdgf1, and Zfp42. In case of Tdgf1, the library targeted the 40 kb region proximal to the gene in an unbiased manner. In other cases, we selected proximal regions to the gene most likely to be involved in regulation based on enhancer-like features that are a maximum of ˜150 kb away from the gene. Additionally, distal regions up to 92 mB away from the gene were added when ChIA-PET distal interaction data⁹ suggested a possible interaction with the target gene promoter using the Sprout algorithm³⁵. In FIGS. 2A-D and 3A-C, the bulk density panel shows the distribution of integrated guide RNAs along the region probed. Among the 3621 gRNAs found to be integrated in at least 1 replicate of Tdgf1, 99% were within 140 bp of the nearest adjacent gRNA and 95% were within 32 bp of the adjacent gRNA. Distances were measured between the mid-points of adjacent gRNAs. The mean distance between adjacent gRNAs was 11 bp, and the standard deviation of this distance was 32 bp. Of note, repetitive and unmappable genomic regions cannot be tiled with gRNAs, and gRNAs targeting regions whose sequence differs from that of the reference genome cannot be appropriately tiled without genome sequence data of the cell line. Each library also contained 10 positive control gRNAs targeting the GFP open reading frame that we expected would cause GFP loss.

We performed four biological replicate screens for Zfp42 and Tdgf1, two replicates for Nanog and a single replicate for Rpp25. Selected screen hits were independently confirmed as described below. Starting one week after electroporation, we collected genomic DNA of the unsorted library-integrated cells to examine differences in gRNA integration. Over 90% of correctly synthesized gRNAs were detected in the genomic DNA for both Tdgf1 and Zfp42 libraries. In addition, gRNA integration rates in biological replicates showed concordance (FIGS. 1C-1D, and data not shown). All of the regulatory regions that we surveyed had adequate coverage of gRNAs to assay their detailed function (Bulk density track, FIGS. 2A-D, 3A-C, data not shown).

Library-integrated mESCs were then flow cytometrically sorted to identify gRNAs inducing loss of GFP expression. Separate GFP^neg and GFP^medium populations were sorted in the Tdgf1^GFP and Zfp42^GFP experiments, while GFP^neg and GFP^medium populations were combined in the Nanog^GFP and Rpp25^GFP experiments because of incomplete population separation (FIG. 1B, data not shown). All sorted populations were >90% pure except the Tdgf1^GFP GFP^medium population, which showed a subpopulation of GFP^neg cells even after multiple sorts, indicating intrinsic fluctuation in GFP expression in this population (data not shown). Integrated gRNAs in the genomic DNA from bulk unsorted as well as sorted populations were deep sequenced to reveal the relative importance of each gRNA-targeted cis-regulatory sequence in governing gene expression (FIGS. 4A-4D, 5A-5D).

The distribution of gRNA abundance in GFP^neg and GFP^medium populations in all screens clearly indicates that a subset of cis-regulatory genomic space is required for gene expression at all four gene loci (FIGS. 2A-2B, FIGS. 3A-C). Importantly, we detected significant overrepresentation of nearly all integrated positive control GFP coding region targeting gRNAs in all replicates (FIG. 2D, FIG. 3C, data not shown) suggesting that MERA robustly identifies gRNAs inducing loss of gene expression. Using the relative abundances of GFP ORF-targeting positive control gRNAs and the dummy gRNA as a negative control, we devised a method to detect gRNAs with statistically significant overrepresentation in GFP^neg and GFP^medium populations (data not shown).

In our MERA screen of Tdgf1 we observed differential enrichment of gRNAs in established functional categories of genomic elements associated with gene regulation (FIGS. 2A, 2D, data not shown). Among these categories were enhancers predicted using chromatin modifications^{32, 33} and classified as strong or weak based on the enrichment of H3K27ac, DNase-I hotspots³⁴ predicted from DNase-seq data³⁵ and transcription factor binding sites predicted from ChIP-seq data³⁶ for 20 transcription factors known to be important in ES cell biology. The highest density of significant gRNAs in the genomic regions were observed at the promoter region for Tdgf1, the neighboring DNase-I hotspot 4 kb upstream adjacent to the Lrrc2 promoter, and the strong enhancer overlapping the Lrrc2 promoter (FIGS. 2A, 2D). We observed other expected trends such as greater enrichment of significant gRNA at predicted strong enhancers proximal to the gene as compared to weak enhancers or enhancers further away.

Surprisingly we observed a novel class of genomic elements downstream of Tdgf1 (FIG. 2A, highlighted in grey) which did not coincide with any known markers of regulatory activity such as H3K27ac, H3K4me1, H3K4me3, known TF binding sites, DNase-I hypersensitivity, predicted DNase-I hotspots, or enhancers predicted from chromatin modifications. We designated such elements that do not contain these markers as Unmarked Regulatory Elements (UREs). Unmarked regulatory regions produced comparable loss of GFP as some distant enhancers (FIG. 2D).

In our MERA screen of Zfp42, we also observed the strongest enrichment for GFP loss in the promoter and proximal enhancer regions (FIGS. 3A, 3C). We observed enrichment of gRNAs in GFP^neg and GFP^medium population at UREs in region II and VII, and also in intervening unmarked regions in regions III to VI (FIG. 3A, data not shown). We also observed the participation of a neighboring promoter Trim12 in the regulation of the Zfp42 gene (FIG. 3A, data not shown). Upon examining the fractional enrichment of gRNAs causing GFP loss in various functional categories, we observe that trends for higher enrichment of gRNAs causing GFP loss are strongly associated with proximity to the gene rather than functional annotation of the region (FIGS. 3A, 3C). gRNAs in UREs flanking enhancers as in regions III to VI are more likely to induce GFP loss than those in distal UREs. Some of these UREs are over 1 kb along as in the case of region IV (chr8:43322062-43323500), hence it is unlikely that GFP loss is an artifact of long disruptions caused by individual gRNAs in the URE region. We also note that regulatory regions upstream of Zfp42 tended to cause intermediate loss of GFP as compared to a complete loss of GFP (GFP^medium vs. GFP^neg, FIG. 3C), suggesting that these enhancers are each responsible for only part of the overall Zfp42 expression level in cells.

To determine the accuracy of the MERA screen in systematically determining required cis-regulatory regions, we first examined replicate consistency among our Tdgf1, Zfp42, and Nanog MERA data. Spatial patterns of gRNA enrichment in the GFP^neg population were largely conserved between replicates (FIG. 2B, FIG. 3B, data not shown). The Pearson correlation between the fraction of GFP^neg gRNA enriched within bins improves with bin length achieving a correlation coefficient of 0.9 at a bin size of 1 kb in Tdgf1. In Zfp42, we observe a correlation of 0.85 even at bin sizes of 100 bp. Thus, MERA has utility in detecting the relative importance of regulatory regions for a particular gene. At an individual level, the overlap between gRNAs enriched in GFP^neg populations between replicates was significant for all replicates in the 3 lines, Tdgf1^GFP, Zfp42^GFP and Nanog^GFP (hyper geometric p-value <0.001); however, some gRNAs found to be significant in one replicate were not observed to be so in another. We found that several of the gRNAs that were significant in only one replicate induced significant GFP loss in an individual test (FIG. 2B), suggesting that the high complexity of MERA data in which a single gRNA can cause hundreds of distinct mutant genotypes leads to some experimental variability.

To analyze potential false positives caused by off-target effects, we built a model of CRISPR off-target cutting using data from 13 gRNAs that were generated by GUIDE-Seq³⁷. We found that in inverse proportion to the GC-content, guide RNAs could tolerate between a maximum of 3 to 6 total mismatches including the PAM region, with up to 3 mismatches in the seed region (9-20 bp) (data not shown). We defined a true negative set as pairs of gRNAs causing significant GFP-loss in at least 2 replicates and gRNAs that were integrated in all replicates but did not cause any significant loss of GFP. Any gRNA found to be non-significant could not possibly have an off-target effect at a site where a significant gRNA was cutting. Using our model of off-target effects no pairs in the above-defined negative set had any predicted interactions (data not shown).

We then used our model of potential off-target effects to determine the promiscuity of gRNAs in the Tdgf1 and Zfp42 libraries. In the Tdgf1 library, 1160/3621 of the integrated gRNAs have potential off-target effects, and 150/925 of the gRNAs that were significantly enriched in GFP^neg populations have one or two potential off-target sites within the topological domain containing the Tdgf1 gene as determined from mESC HiC data³⁸. In the Zfp42 library, 632/1643 integrated guide RNAs have predicted off-target effects, and 34/332 of the gRNAs enriched in GFP^neg cells have predicted off-target effects in the topological domain containing the Zfp42 gene³⁸. We found that when we eliminate gRNAs with potential off-target effects from our analysis, the global distribution of significantly enriched gRNAs along the regulatory landscape of the gene is unaltered and relative contributions of different functional categories are unaffected (data not shown). Furthermore, several gRNAs with no predicted off-target effects support the regulation of Tdgf1 by the promoter of Lrrc2 (data not shown), the promoter of Trim12, and a URE region (data not shown), and none of these regions are more likely to contain off-target effects than other screened regions. Our off-target predictions are overly cautious, as off-target cutting is typically much more rare than on-target cutting³⁷ and the off-target sites predicted for MERA hits are often >200 kb away from the gene reducing the likelihood of there existing a functional association. However, even this overestimate of off-target effects does not alter the patterns seen in MERA data.

To analyze potential off-target effects with an independent method, we asked whether any gRNAs from the Tdgf1 library would extinguish Zfp42-GFP activity and vice versa. We found that a much smaller percentage of cells lose GFP upon targeting by a mismatched gRNA library than by the matched library (data not shown). Sequencing revealed that the gRNAs enriched in GFP^neg mismatched library-targeted cells were predominantly GFP control gRNAs with a small number of non-clustered gRNAs displaying off-target activity (data not shown). Thus, the clustered enrichment of GFP loss at enhancers, neighboring promoters, and UREs in MERA is not replicated by computationally predicted or experimentally determined off-target effects, leading us to conclude that GFP loss in these regions is a result off on-target gRNA effects (data not shown).

To determine the false positive rate at the level of individual gRNAs, we introduced individual gRNAs to determine whether their rate of GFP loss correlated with their activity in the pooled MERA screen. These gRNAs fell within several of the functional categories including UREs and neighboring promoters (FIG. 2A highlighted in grey, FIG. 2B). We confirmed significantly increased GFP loss in 29/30 gRNAs from these screens as compared to five similarly located control gRNAs (FIG. 2B). Nine of these gRNA were in the URE regions downstream of the Tdgf1. Altogether, we conclude that MERA has a low false positive rate.

We next sought to determine the false negative rate of MERA. We found 10/10 GFP-targeting gRNAs in all 4 GFP-lines are highly enriched in GFP^neg cells (FIG. 2D, FIG. 3C). Additionally, 67/83 (81%) gRNAs that target the first 700 bp of the Rpp25 open-reading frame are highly enriched in GFP^neg cells. In a 500 bp around the Tdgf1 promoter region, 48/59 (81%) of gRNAs induce GFP loss in multiple replicates (data not shown). Thus, we find that a high percentage of gRNAs expected to have an effect on gene expression are enriched in GFP^neg cells. It is unclear whether the 20% of gRNAs in these regions that do not induce GFP loss are false negatives or true negatives, as their mechanism of inducing GFP loss is not as direct as when the GFP ORF itself is targeted. However, even if this appreciable percentage of individual gRNAs are false negatives, it does not impair the ability of MERA to determine required regulatory regions, as the high density of gRNAs in a region (˜1 per 8 bp) allows highly reproducible resolution at the level of 100-1000 bp (FIGS. 2C, 3B).

To follow up on a specific region in which only some gRNAs induce GFP loss, we tested two pairs of adjacent gRNAs in which one gRNA induces GFP loss whereas the other does not. In one pair we discovered that the gRNA that did not induce loss of GFP was able to cut DNA efficiently and thus was a true negative gRNA (data not shown). In the other pair, we found that the gRNA that did not induce GFP loss was impaired at cutting DNA because a SNP in the target genome changed an expected PAM sequence from NGG to NAG (data not shown). While this analysis is low-throughput, it illustrates that false negatives at the level of individual gRNAs can be caused by differences between the reference genotype and the genotype of the cell line used in our study.

Since each gRNA induces a spectrum of mutations at its target site, this genotypic diversity induced by the same gRNA can result in a range of GFP expression after introduction of each gRNA. For instance, individual targeting of some gRNAs in promoter regions induced GFP loss in over 40% of cells, whereas gRNAs targeting other promoter-distal elements induced GFP loss in only 5-10% of cells (FIG. 2B). Targeting the GFP ORF induces GFP loss in >90% of cells, and negative controls induce loss in <2% of cells in these assays. This is in stark contrast to CRISPR/Cas9-based gene inactivation screens in which all gRNAs are assumed to be equivalently likely to induce frameshift mutations that inactivate gene function. We utilize the genotypic heterogeneity for functional motif discovery as described herein. Regarding false negatives, however, we conclude that MERA has a low but appreciable false negative rate attributable primarily to the large range of mutations that each gRNA can induce and the variable likelihood that mutations at each site will induce GFP loss.

Our MERA results revealed that Tdgf1, Nanog, Rpp25 and Zfp42 have different regulatory architectures (FIGS. 2A-D, FIGS. 3A-C, data not shown). All regulatory regions within +/−20 kb of the Nanog promoter were associated with clusters of highly enriched gRNAs, and 20% to 40% of the tested gRNAs in predicted enhancers and DNase-I hotspots proximal to Nanog resulted in GFP′ cells (data not shown). In contrast, the Rpp25 gene shows a dense concentration of significant gRNAs at its promoter and short ORF region. Other proximal regulatory regions of Rpp25 had 12% of tested gRNAs resulting in GFP^neg cells (data not shown). Tdgf1 shows a similar trend to Nanog with dense clusters of significant gRNA in the proximal regulatory regions (FIGS. 2A, 2D). UREs were also seen in cis-regulatory regions near Rpp25 (data not shown). In Nanog, a distal ChIA-Pet region >92 mB away showed several strongly enriched gRNAs (data not shown), indicating that MERA is capable of measuring the functionality of long-distance chromatin interactions. One observation common to all genes is the participation of the promoters of other genes in regulation. In some cases these gene promoters are several million bases away. Examples of foreign promoter involvement can be seen in the case of Lrrc2 promoter in Tdgf1 (FIGS. 2A, 2D), Mirc35hg in Nanog (data not shown), Scamp5 and Cox5a in Rpp25 (data not shown). Previous studies have documented the existence of dual property elements³⁹ that can act as either promoter or enhancer in different cellular contexts. Additionally, it is known that neighboring promoters often interact with each other⁴⁰ and that neighboring gene expression is often coordinated⁴¹. Here we observe that active promoters may coordinate gene expression patterns of neighboring genes by functioning as enhancers within the same cellular context.

The second phase of MERA uses functional motif discovery to identify the causal elements governing expression at MERA screen hits. Because Cas9 induces random mutations, a pool of mESCs treated with Cas9 and a single gRNA will contain thousands of distinct mutant genotypes centered on the gRNA cleavage site. Recently, TAL effector nucleases have been used to derive functional footprints of regulatory DNA⁴². We hypothesized that we could pinpoint DNA sequence motif(s) that cause GFP loss by identifying sequence features that consistently differ between thousands of GFP^pos and GFP^neg genotypes at a given site (FIG. 4A). Functional motif discovery proceeds by performing individual scCRISPR-mediated mutation by a selected gRNA and obtaining thousands of genotypes from both GFP^pos and GFP^medium/neg cells by high-throughput sequencing and then summarizing the observed genotypes as motifs that reveal what bases are important for gene expression (FIG. 4A). Using the differences in fractions of genotypes at positions along the gRNA, we defined a base-level importance score that was independent of the cutting biases of the gRNA. We computed the Hellinger score of the distance of the mutational profile at each base (4 base possibilities plus a deletion) to the reference base for both GFP^pos and GFP^neg populations. The log-ratio of distance of GFP^neg to the reference to the distance of GFP^pos to the reference provided a base-level estimate of the importance of the base in causing GFP loss.

We first tested to see if functional motif discovery in Tdgf1 and Zfp42 enhancer regions would permit us to classify held out genotypes as GFP^neg or GFP^pos and identify motifs for transcription factors that are directly bound to these regions in mESC cells. We selected two overlapping gRNAs for functional motif discovery in a Tdgf1 proximal enhancer that overlapped binding sites for the key mESC transcription factors Stat3, Sox2 and Tcfcp2lI, of which Stat3 is the only factor with a direct binding site. We were able to classify held out genotypes with an AUC of 0.81 (FIG. 4C), and observed an enrichment of the bases for the Stat3 motif²⁹ in both the left and right paired end reads (FIG. 4D, data not shown). In a Zfp42 enhancer, we selected two gRNAs separated by ˜100 bp with the first gRNA target containing a site for Nrf1 binding (TGCGCAG) and the second gRNA containing a direct binding site for Smad3/Med1/K14 and an indirect binding site for p300 (data not shown). Functional motif discovery for these Zfp42 gRNAs classified held out genotypes with AUCs of 0.77 and 0.71 (data not shown) and revealed bases near the Nrf1 site, around the p300 site, and a slight enrichment of the G-rich tract proximal to the Klf4 binding motif (data not shown).

We next applied functional motif discovery to 2 gRNAs that are in a URE ˜12 kb downstream of the Tdgf1 transcript (FIG. 5A). We observed higher enrichment of mutated bases in GFP^neg genotypes over GFP^pos genotypes for both gRNAs (FIG. 5C, data not shown), and we found distinctive patterns of base importance scores emerging for both gRNAs (FIG. 5D, data not shown). We found that we could obtain high classification accuracy for held out genotypes from both gRNAs (AUC 0.81 and 0.76) for separating GFP^pos and GFP^neg populations using mutations within −20 to +20 bp as features (FIG. 5B, data not shown). It was also noted that for both these gRNAs, the majority of mutations were contained within 20 bp of the gRNA (data not shown). Insertions were enriched in the GFP^neg population vs. GFP^pos but they were less common than deletions (data not shown). Altogether, we conclude that functional motif discovery is a valuable method for ascertaining which bases at MERA-identified regulatory regions are required for gene expression. In enhancer regions, these bases correspond to known binding motifs, and in UREs, we identify blocks of bases which are required for gene expression.

We then used homologous recombination to validate that the Tdgf1 enhancer and URE regulatory elements are truly required for gene expression in the third phase of MERA. We used flanking gRNAs to induce short (>100 bp) deletions in two regions predicted to induce GFP loss by our MERA screen, one in the Tdgf1 enhancer and one at a URE. As expected, a subset of cells lost GFP expression, and we obtained clonal GFP^neg lines containing the deletion genotype (FIGS. 6A, 6B). We then used homology-directed repair to restore the wildtype genotype in these cells, finding at each site that a large percentage of cells reverted to a GFP^pos state (FIG. 6C). We replicated this experiment in wildtype cells without a Tdgf1-GFP allele, finding that clonal deletion cells lost Tdgf1 RNA expression, and clonal repaired lines restored Tdgf1 expression (FIGS. 6A-6C). This robust and straightforward relationship between local genotype and GFP expression provides compelling evidence that the local DNA sequence at a URE is required for Tdgf1 expression.

In conclusion, MERA offers a new unbiased, high-resolution approach to directly interrogate the regulatory genome for function. MERA not only provides a survey of required cis-regulatory elements, it also enables functional motif discovery to dissect the precise nature of identified regulatory elements.

MERA enabled us to discover unmarked regulatory elements (UREs) that are not associated with conventionally expected DNase hypersensitivity and histone mark features. This observation reinforces the importance of direct perturbation analysis to definitively characterize genome function, as we observe that correlative analysis is insufficient. While we do not yet have definitive data as to the function of UREs, we find that a URE downstream of the Tdgf1 gene is highly sensitive to base substitution at a string of consecutive bases, suggesting that its DNA sequence is crucial to its regulatory activity. Further, we find the first half of this URE to be highly conserved (phastcons score>0.85, data not shown) indicating potential functional significance of the genomic region. Consistent with these data, UREs may be RNA templates (data not shown), elements bound by uncharacterized protein factors, or spacers where their precise base sequence is of secondary importance.

We designed our gRNA libraries to target a mix of previously annotated and unannotated cis-regulatory regions, and thus we did not uniformly tile the proximal regions of any of these genes. Therefore, we cannot estimate the frequency of UREs and expect that future MERA screens with even more extensive coverage at more loci will elucidate how pervasive UREs are in the regulatory architecture of the genome.

MERA is complementary to high-throughput reporter assays, which assess elements sufficient to induce gene expression, and future experiments performing both approaches on a defined cohort of regions should provide insight into the degree of concordance between necessary and sufficient gene regulatory elements. MERA also enables quantitative assessment of the relative contributions of distinct cis-regulatory elements on gene expression, and future studies will provide important insights into how regulatory regions combine to achieve desired levels of expression. Lastly, extending MERA to explore how changes in individual cis-regulatory elements alter gene networks will aid our understanding of how cis-regulatory variants lead to human disease. We expect that the direct interrogation of variant locations discovered in genome wide association studies by MERA will provide a rapid way to screen such variants for function in relevant cell types.

REFERENCES

• 1. Jenuwein, T. & Allis, C. D. Translating the histone code. Science 293, 1074-1080 (2001).
• 2. Bernstein, B. E. et al. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125, 315-326 (2006).
• 3. Rada-Iglesias, A. et al. A unique chromatin signature uncovers early developmental enhancers in humans. Nature 470, 279-283 (2011).
• 4. Heintzman, N. D. et al. Histone modifications at human enhancers reflect global cell-type-specific gene expression. Nature 459, 108-112 (2009).
• 5. Creyghton, M. P. et al. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proceedings of the National Academy of Sciences of the United States of America 107, 21931-21936 (2010).
• 6. Melnikov, A. et al. Systematic dissection and optimization of inducible enhancers in human cells using a massively parallel reporter assay. Nature biotechnology 30, 271-277 (2012).
• 7. Arnold, C. D. et al. Genome-wide quantitative enhancer activity maps identified by STARR-seq. Science 339, 1074-1077 (2013).
• 8. Patwardhan, R. P. et al. Massively parallel functional dissection of mammalian enhancers in vivo. Nature biotechnology 30, 265-270 (2012).
• 9. Fullwood, M. J. et al. An oestrogen-receptor-alpha-bound human chromatin interactome. Nature 462, 58-64 (2009).
• 10. Lieberman-Aiden, E. et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289-293 (2009).
• 11. Simonis, M. et al. Nuclear organization of active and inactive chromatin domains uncovered by chromosome conformation capture-on-chip (4C). Nature genetics 38, 1348-1354 (2006).
• 12. Dostie, J. et al. Chromosome Conformation Capture Carbon Copy (5C): a massively parallel solution for mapping interactions between genomic elements. Genome research 16, 1299-1309 (2006).
• 13. Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816-821 (2012).
• 14. Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819-823 (2013).
• 15. Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823-826 (2013).
• 16. Jinek, M. et al. RNA-programmed genome editing in human cells. eLife 2, e00471 (2013).
• 17. Fu, Y., Sander, J. D., Reyon, D., Cascio, V. M. & Joung, J. K. Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. Nature biotechnology 32, 279-284 (2014).
• 18. Cho, S. W. et al. Analysis of off-target effects of CRISPR/Cas-derived RNA-guided endonucleases and nickases. Genome research 24, 132-141 (2014).
• 19. Gilbert, L. A. et al. CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442-451 (2013).
• 20. Fu, Y. et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nature biotechnology 31, 822-826 (2013).
• 21. Kleinstiver, B. P. et al. Engineered CRISPR-Cas9 nucleases with altered PAM specificities. Nature 523, 481-485 (2015).
• 22. Cradick, T. J., Fine, E. J., Antico, C. J. & Bao, G. CRISPR/Cas9 systems targeting beta-globin and CCR5 genes have substantial off-target activity. Nucleic acids research 41, 9584-9592 (2013).
• 23. Wang, T., Wei, J. J., Sabatini, D. M. & Lander, E. S. Genetic screens in human cells using the CRISPR-Cas9 system. Science 343, 80-84 (2014).
• 24. Shalem, O. et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343, 84-87 (2014).
• 25. Zhou, Y. et al. High-throughput screening of a CRISPR/Cas9 library for functional genomics in human cells. Nature 509, 487-491 (2014).
• 26. Koike-Yusa, H., Li, Y., Tan, E. P., Velasco-Herrera Mdel, C. & Yusa, K. Genome-wide recessive genetic screening in mammalian cells with a lentiviral CRISPR-guide RNA library. Nature biotechnology 32, 267-273 (2014).
• 27. Chen, S. et al. Genome-wide CRISPR screen in a mouse model of tumor growth and metastasis. Cell 160, 1246-1260 (2015).
• 28. Orlando, S. J. et al. Zinc-finger nuclease-driven targeted integration into mammalian genomes using donors with limited chromosomal homology. Nucleic acids research 38, e152 (2010).
• 29. Young, R. A. Control of the embryonic stem cell state. Cell 144, 940-954 (2011).
• 30. Yue, F. et al. A comparative encyclopedia of DNA elements in the mouse genome. Nature 515, 355-364 (2014).
• 31. Singh, A. M., Hamazaki, T., Hankowski, K. E. & Terada, N. A heterogeneous expression pattern for Nanog in embryonic stem cells. Stem cells 25, 2534-2542 (2007).
• 32. Rajagopal, N. et al. RFECS: a random-forest based algorithm for enhancer identification from chromatin state. PLoS computational biology 9, e1002968 (2013).
• 33. Consortium, E. P. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57-74 (2012).
• 34. John, S. et al. Chromatin accessibility pre-determines glucocorticoid receptor binding patterns. Nature genetics 43, 264-268 (2011).
• 35. Sherwood, R. I. et al. Discovery of directional and nondirectional pioneer transcription factors by modeling DNase profile magnitude and shape. Nature biotechnology 32, 171-178 (2014).
• 36. Guo, Y., Mahony, S. & Gifford, D. K. High resolution genome wide binding event finding and motif discovery reveals transcription factor spatial binding constraints. PLoS computational biology 8, e1002638 (2012).
• 37. Tsai, S. Q. et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nature biotechnology 33, 187-197 (2015).
• 38. Dixon, J. R. et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature 485, 376-380 (2012).
• 39. Leung, D. et al. Integrative analysis of haplotype-resolved epigenomes across human tissues. Nature 518, 350-354 (2015).
• 40. Li, G. et al. Extensive promoter-centered chromatin interactions provide a topological basis for transcription regulation. Cell 148, 84-98 (2012).
• 41. Woo, Y. H., Walker, M. & Churchill, G. A. Coordinated expression domains in mammalian genomes. PloS one 5, e12158 (2010).
• 42. Vierstra, J. et al. Functional footprinting of regulatory DNA. Nature methods (2015).

EQUIVALENTS

While several inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and/or structures for performing the function and/or obtaining the results and/or one or more of the advantages described herein, and each of such variations and/or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and/or configurations will depend upon the specific application or applications for which the inventive teachings is/are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and/or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and/or methods, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and/or ordinary meanings of the defined terms.

All references, patents and patent applications disclosed herein are incorporated by reference with respect to the subject matter for which each is cited, which in some cases may encompass the entirety of the document.

The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

The phrase “and/or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and/or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and/or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and/or” as defined above. For example, when separating items in a list, “or” or “and/or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of” or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e. “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and/or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

It should also be understood that, unless clearly indicated to the contrary, in any methods claimed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited.

In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of” and “consisting essentially of” shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.

Claims (14)

What is claimed is:

1. A multiplexed editing regulatory assay (MERA) construct comprising: one or more copies of a defective gRNA expression construct wherein each defective gRNA expression construct comprises:

i. a dummy gRNA that is not homologous to a eukaryotic genome, and is unable to target a Cas nuclease to the genome for a double strand break,

ii. a guide hairpin sequence,

iii. a native regulatory region which will be targeted using CRISPR/Cas endonuclease; and

iv. an operably linked promoter.

2. The MERA construct of claim 1, wherein the eukaryotic genome is a human genome.

3. The MERA construct of claim 1, wherein the gRNA is 19-21 nucleotides in length.

4. The MERA construct of claim 1, wherein the guide hairpin sequence is about 40 nucleotides in length and once transcribed can be bound to a CRISPR/Cas nuclease.

5. A host cell comprising the MERA construct of claim 1.

6. The host cell of claim 5, wherein a single defective gRNA expression construct is integrated into the host cell genome.

7. The host cell of claim 5, wherein 2-10 defective gRNA expression constructs are integrated into the host cell genome.

8. A population of host cells of claim 5.

9. The host cell of claim 5, further comprising a CRISPR/Cas nuclease and/or a coding sequence for the CRISPR/Cas nuclease.

10. The host cell of claim 5, further comprising Cas9 nuclease.

11. The host cell of claim 9 or 10, further comprising a library of exogenous gRNA homology fragments spanning across the targeted native regulatory region, optionally wherein the gRNA homology fragment further comprises a coding sequence for a reporter protein or a reporter fusion protein.

12. A population of cells of claim 11, wherein each cell has integrated into its genome one or more functional gRNA expression construct at the dummy guide locus.

13. The population of cells of claim 12, wherein the genomic integration is in a virus-independent manner.

14. A method comprising:

(1) introducing into a host cell comprising the MERA construct of claim 1: a library of exogenous gRNA homology fragments spanning across the targeted native regulatory region, wherein the gRNA homology fragment further comprises a coding sequence for a reporter protein or a reporter fusion protein, and

(2) allowing a single exogenous gRNA homology fragment to homologously recombine with the dummy gRNA construct in the presence of a CRISPR/Cas nuclease to form a functional guide RNA construct, wherein the host cell comprises a reporter gene coding sequence integrated into its genome.

Images